<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Bangaly Kaba on Medium]]></title>
        <description><![CDATA[Stories by Bangaly Kaba on Medium]]></description>
        <link>https://medium.com/@fbkaba?source=rss-c5c68eb01121------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*0j5IHSdjYBXtytEA.jpg</url>
            <title>Stories by Bangaly Kaba on Medium</title>
            <link>https://medium.com/@fbkaba?source=rss-c5c68eb01121------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 07:11:19 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fbkaba/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Building Speaker-Aware Video Summaries on AWS]]></title>
            <link>https://medium.com/@fbkaba/building-speaker-aware-video-summaries-on-aws-f38662e54513?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/f38662e54513</guid>
            <category><![CDATA[aws-titan]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[claude-haiku]]></category>
            <category><![CDATA[aws-transcribe]]></category>
            <category><![CDATA[bedrock]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Sun, 18 Jan 2026 21:45:16 GMT</pubDate>
            <atom:updated>2026-01-18T21:52:45.148Z</atom:updated>
            <content:encoded><![CDATA[<h3>Hands-On with Amazon Transcribe and Amazon Bedrock</h3><p><strong>Bangaly Kaba</strong><br>Cloud, Security &amp; Generative AI Architect</p><h3>Introduction</h3><p>In <strong>Part 1</strong>, we explored the architectural foundations for multimodal video summarization on AWS and explained why <strong>architecture matters more than the model</strong> when operationalizing Generative AI.</p><p>In this article, we move from design to <strong>implementation</strong>.</p><p>The objective is deliberately practical:</p><p><strong>Starting from a recorded video, generate a speaker-aware summary that identifies who spoke and what they contributed — using AWS managed services.</strong></p><p>This pattern applies directly to conference recordings, meetups, training sessions, and internal knowledge platforms.</p><h3>From Media to Insights: Processing Pipeline</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*eVW5iwKhx_MVpzFS.png" /><figcaption>AWS reference architecture for speaker-aware video summarization using Amazon Bedrock</figcaption></figure><p>At a high level, the pipeline consists of three clear steps:</p><ol><li><strong>Speaker-aware transcription</strong></li><li><strong>LLM-based reasoning and summarization</strong></li><li><strong>Structured output for reuse</strong></li></ol><p>Each step is handled by a <strong>purpose-built AWS service</strong>, keeping the system modular and evolvable.</p><h3>Step 1 — Speaker Diarization with Amazon Transcribe</h3><p>The first requirement is to convert audio into text <strong>while preserving speaker context</strong>.</p><p><strong>Amazon Transcribe</strong> supports speaker diarization out of the box, producing transcripts labeled by speaker identity.</p><h3>Example: Starting a Transcription Job</h3><pre>import boto3<br><br>transcribe = boto3.client(&quot;transcribe&quot;, region_name=&quot;us-east-1&quot;)<br>transcribe.start_transcription_job(<br>    TranscriptionJobName=&quot;speaker-aware-video-summary&quot;,<br>    Media={&quot;MediaFileUri&quot;: &quot;s3://my-bucket/video-audio.mp3&quot;},<br>    MediaFormat=&quot;mp3&quot;,<br>    LanguageCode=&quot;en-US&quot;,<br>    Settings={<br>        &quot;ShowSpeakerLabels&quot;: True,<br>        &quot;MaxSpeakerLabels&quot;: 5<br>    }<br>)</pre><h3>Resulting Transcript (Excerpt)</h3><pre>[SPEAKER_0]: Welcome everyone to today’s session.<br>[SPEAKER_1]: I’ll walk through the architecture decisions.</pre><p>At this stage, we are no longer dealing with raw text, but with <strong>structured conversational data</strong> — a critical prerequisite for meaningful reasoning.</p><h3>Step 2 — Prompt Design as an Architectural Decision</h3><p>Before invoking a foundation model, it is essential to define <strong>what the model is responsible for — and what it is not</strong>.</p><p>Prompt design is therefore part of the architecture.</p><h3>Prompt Goals</h3><ul><li>Identify active speakers</li><li>Summarize each speaker’s contribution</li><li>Produce consistent, neutral output</li></ul><h3>Example Prompt</h3><pre>You are an AI content analyzer.<br><br>Given the transcript below with speaker labels:<br>- Identify active speakers<br>- Summarize each speaker&#39;s contribution<br>- Indicate relative participation level<br>Return a clear, structured summary.</pre><p>Clear constraints lead to <strong>predictable outputs</strong>, lower cost, and reduced hallucination.</p><h3>Step 3 — Reasoning with Amazon Bedrock</h3><p>The reasoning step is handled using <strong>Amazon Bedrock</strong>.</p><p>Amazon Bedrock allows you to:</p><ul><li>Invoke foundation models without managing infrastructure</li><li>Switch models without redesigning the system</li><li>Integrate directly with AWS security and IAM</li></ul><h3>Bedrock Invocation Example (Claude 3)</h3><pre>import boto3<br>import json<br><br>bedrock = boto3.client(&quot;bedrock-runtime&quot;, region_name=&quot;us-east-1&quot;)<br>prompt_with_transcript = &quot;&quot;&quot;<br>You are an AI content analyzer.<br>Transcript:<br>[SPEAKER_0]: Welcome everyone to today&#39;s session.<br>[SPEAKER_1]: I will present the architecture and key design choices.<br>&quot;&quot;&quot;<br>body = {<br>    &quot;anthropic_version&quot;: &quot;bedrock-2023-05-31&quot;,<br>    &quot;messages&quot;: [<br>        {<br>            &quot;role&quot;: &quot;user&quot;,<br>            &quot;content&quot;: prompt_with_transcript<br>        }<br>    ],<br>    &quot;max_tokens&quot;: 800,<br>    &quot;temperature&quot;: 0.3<br>}<br>response = bedrock.invoke_model(<br>    modelId=&quot;anthropic.claude-3-sonnet-20240229-v1:0&quot;,<br>    body=json.dumps(body)<br>)<br>result = json.loads(response[&quot;body&quot;].read())<br>print(result[&quot;content&quot;][0][&quot;text&quot;])</pre><h3>Sample Output</h3><pre>Speaker Summary:<br><br>- SPEAKER_0<br>  Role: Host<br>  Contribution: Introduced the session and guided the discussion<br>  Participation Level: Medium<br>- SPEAKER_1<br>  Role: Presenter<br>  Contribution: Explained architecture and key design decisions<br>  Participation Level: High</pre><p>This output is:</p><ul><li>Human-readable</li><li>Easy to store</li><li>Easy to convert to JSON for dashboards or search</li></ul><h3>Model Selection Considerations</h3><p>One advantage of Amazon Bedrock is <strong>model flexibility</strong>.</p><p>ModelTypical UseClaude 3 SonnetLong reasoning, rich summariesClaude 3 HaikuFast, cost-efficient summariesAmazon Titan TextAWS-native text workloads</p><p>Model choice becomes a <strong>runtime decision</strong>, not a redesign.</p><h3>Orchestration and Persistence</h3><p>The workflow is orchestrated using <strong>AWS Lambda</strong>:</p><ul><li>Triggered by S3 uploads or Transcribe completion events</li><li>Stateless and cost-efficient</li><li>Easy to evolve as requirements change</li></ul><p>Summaries can be stored in:</p><ul><li>Amazon S3 for durability</li><li>DynamoDB for indexed access</li><li>Search systems for discovery</li></ul><h3>Why This Pattern Works</h3><p>This approach succeeds because it:</p><ul><li>Separates speech recognition from reasoning</li><li>Keeps foundation models focused on what they do best</li><li>Aligns with AWS Well-Architected principles</li><li>Scales from community meetups to enterprise workloads</li></ul><p>Most importantly, it avoids treating Generative AI as a monolith.</p><h3>What Comes Next</h3><p>This article focused on <strong>implementation mechanics</strong>.</p><p>In <strong>Part 3</strong>, we address the most frequently underestimated aspects of Generative AI:</p><ul><li>Cost control</li><li>IAM and access boundaries</li><li>Observability</li><li>Responsible AI practices</li></ul><p>These elements determine whether a GenAI system can <strong>survive in production</strong>.</p><h3>Conclusion</h3><p>Speaker-aware video summarization is a realistic and high-value Generative AI use case.</p><p>By combining <strong>Amazon Transcribe</strong> and <strong>Amazon Bedrock</strong>, AWS enables builders to move from transcripts to insights while maintaining security, scalability, and operational discipline.</p><p><strong>Generative AI delivers value when it is engineered — not improvised.</strong></p><p><em>Part 3: Operating Generative AI on AWS — Cost, Security, and Responsible AI</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f38662e54513" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Operational Multimodal Video Summarization on AWS]]></title>
            <link>https://medium.com/@fbkaba/operational-multimodal-video-summarization-on-aws-a6bd1acf9eff?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/a6bd1acf9eff</guid>
            <category><![CDATA[generative-ai-use-cases]]></category>
            <category><![CDATA[vector-embeddings]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[bedrock]]></category>
            <category><![CDATA[knowledge-base]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Sun, 18 Jan 2026 21:29:30 GMT</pubDate>
            <atom:updated>2026-01-18T21:29:30.912Z</atom:updated>
            <content:encoded><![CDATA[<h3>Architecture and Design Foundations with Amazon Bedrock</h3><p><strong>Bangaly Kaba</strong><br>Cloud, Security &amp; Generative AI Architect</p><h3>Introduction</h3><p>Video has become one of the primary ways knowledge is shared today. Conferences, meetups, training sessions, and internal presentations are increasingly recorded and consumed asynchronously. While access to content has improved, <strong>extracting meaningful insights from long videos remains difficult</strong>.</p><p>Generative AI changes this equation. Beyond simple transcription, modern foundation models can reason over content, identify speakers, and synthesize structured summaries. However, many demonstrations stop at isolated prompts and overlook a critical question:</p><p><strong>How do we operationalize Generative AI on AWS in a secure, scalable, and reusable way?</strong></p><p>This article introduces the <strong>architectural foundations</strong> for multimodal video summarization on AWS, with <strong>Amazon Bedrock</strong> as the reasoning engine. The focus is intentionally on <strong>design patterns</strong>, not demos.</p><h3>The Real Problem with Video Content</h3><p>Organizations often ask:</p><ul><li>Who actively participated in a recorded session?</li><li>What did each speaker contribute?</li><li>How can summaries be reused across teams and platforms?</li></ul><p>Traditional speech-to-text solves only part of the problem. Transcripts alone are <strong>not insights</strong>. The real value comes from <strong>reasoning over structured signals</strong>, applying context, and generating summaries that humans can quickly consume.</p><p>This is where multimodal and large language models become powerful — <em>if they are integrated correctly</em>.</p><h3>Why Architecture Matters More Than the Model</h3><p>A common mistake in Generative AI projects is to treat the foundation model as the system.</p><p>In practice:</p><ul><li>Models evolve quickly</li><li>APIs change</li><li>Context windows are finite</li><li>Costs scale with usage</li></ul><p>A <strong>well-designed architecture</strong> absorbs this change. The model becomes a component, not the foundation.</p><p>On AWS, this naturally leads to a <strong>pipeline-based design</strong> where each service has a clear responsibility.</p><h3>Core Design Principles</h3><p>Before looking at services, it is important to clarify the principles guiding this architecture.</p><h3>1. Separation of Concerns</h3><p>Speech recognition, reasoning, orchestration, and storage should be independent.</p><h3>2. Serverless by Default</h3><p>No infrastructure to manage. Scale should be automatic.</p><h3>3. Security and Governance First</h3><p>Identity, access control, and data isolation must be explicit.</p><h3>4. Model Flexibility</h3><p>The system should not depend on a single foundation model.</p><p>These principles align closely with the <strong>AWS Well-Architected Framework</strong>.</p><h3>Reference Architecture Overview</h3><p>At a high level, multimodal video summarization on AWS is implemented as a <strong>four-stage pipeline</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/936/0*T-g-qO96r51qNuHZ.png" /><figcaption><strong>Figure — Reference architecture for multimodal video summarization on AWS.</strong><br> Video and audio content is stored in Amazon S3, transcribed with speaker diarization using Amazon Transcribe, analyzed by Amazon Bedrock for reasoning and summarization, and orchestrated using AWS serverless services.</figcaption></figure><h3>Architecture Components</h3><h3>1. Media Storage</h3><p>Video or audio content is stored in <strong>Amazon S3</strong>.</p><p>This provides:</p><ul><li>Durable storage</li><li>Event-driven processing</li><li>Simple integration with downstream services</li></ul><h3>2. Speech-to-Text with Speaker Awareness</h3><p><strong>Amazon Transcribe</strong> converts audio into text while identifying individual speakers.</p><p>The output is no longer raw text, but <strong>structured conversation data</strong>:</p><ul><li>Speaker labels</li><li>Time alignment</li><li>Clear segmentation</li></ul><p>This structure is critical for meaningful reasoning.</p><h3>3. Reasoning and Summarization</h3><p>This is where <strong>Amazon Bedrock</strong> comes into play.</p><p>Rather than embedding logic inside the model call, Bedrock acts as a <strong>managed reasoning layer</strong>, allowing teams to:</p><ul><li>Select the appropriate foundation model (Claude, Titan, Llama)</li><li>Apply consistent prompts</li><li>Enforce security and isolation</li><li>Scale without managing infrastructure</li></ul><p>The model receives <strong>speaker-attributed transcripts</strong>, not raw media, which improves accuracy and reduces cost.</p><h3>4. Orchestration and Integration</h3><p><strong>AWS Lambda</strong> orchestrates the workflow:</p><ul><li>Triggered by S3 events or jobs</li><li>Stateless and cost-efficient</li><li>Easy to evolve as requirements change</li></ul><p>Summaries can then be stored in S3 or indexed for search and analytics.</p><h3>Why Amazon Bedrock Is the Right Fit</h3><p>Amazon Bedrock is particularly well suited for this use case because it provides:</p><ul><li><strong>Model choice without lock-in</strong><br>Change models without redesigning the system.</li><li><strong>Enterprise-grade security</strong><br>IAM integration, private networking, and data isolation.</li><li><strong>Operational simplicity</strong><br>No GPU management, no scaling concerns.</li><li><strong>Cost transparency</strong><br>Pay only for what you use.</li></ul><p>This makes Bedrock ideal not only for enterprises, but also for <strong>community-driven and educational workloads</strong>.</p><h3>From Architecture to Implementation</h3><p>This article intentionally stops at the <strong>design level</strong>.</p><p>In the next parts of this series:</p><ul><li><strong>Part 2</strong> will go hands-on, covering speaker diarization and Bedrock invocation</li><li><strong>Part 3</strong> will focus on cost control, security, observability, and Responsible AI</li></ul><p>The goal is not just to make Generative AI work — but to make it <strong>work well in the real world</strong>.</p><h3>Conclusion</h3><p>Multimodal video summarization is a powerful and realistic Generative AI use case. But success depends far more on <strong>architecture and operational discipline</strong> than on the choice of model.</p><p>By combining <strong>Amazon Transcribe</strong>, <strong>Amazon Bedrock</strong>, and <strong>serverless AWS services</strong>, builders can create <strong>secure, scalable, and reusable</strong> systems that turn raw video into actionable knowledge.</p><p><strong>Generative AI delivers value when it is operational, observable, and responsibly designed.</strong></p><p><em>Part 2: Building the Pipeline — Speaker-Aware Summarization with Amazon Transcribe and Amazon Bedrock</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a6bd1acf9eff" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Building Secure, Intelligent, and Cloud-Native Applications: My Journey With Containers, GenAI…]]></title>
            <link>https://medium.com/@fbkaba/building-secure-intelligent-and-cloud-native-applications-my-journey-with-containers-genai-6b9c3f0cab02?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/6b9c3f0cab02</guid>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Thu, 27 Nov 2025 12:23:59 GMT</pubDate>
            <atom:updated>2025-11-27T12:23:59.085Z</atom:updated>
            <content:encoded><![CDATA[<h3>Building Secure, Intelligent, and Cloud-Native Applications: My Journey With Containers, GenAI, and AWS</h3><p>Technology moves fast, but some shifts truly transform the industry.<br> Working across <strong>cloud, cybersecurity, and banking</strong>, I’ve watched three revolutions reshape how we build applications:</p><ul><li><strong>Containers and Kubernetes</strong></li><li><strong>Cloud-native security and governance</strong></li><li><strong>Generative AI becoming production-ready</strong></li></ul><p>And after years designing systems in highly regulated environments, I’ve learned one thing:</p><p><strong>Containers gave us speed.<br>GenAI gave us intelligence.<br>AWS gives us the structure to do it securely.</strong></p><p>This article is my personal reflection — not theory — on how these pieces fit together, how they changed the way I build, and why I’m committed to mentoring the next generation of cloud professionals.</p><h3>1. Containers gave us agility, but AWS made them safe for the real world</h3><p>When containers became popular, many organizations in finance were excited… and nervous.</p><p>Packaging apps into containers was simple.<br> Running them securely at scale, with the level of governance banks require? Much harder.</p><p>AWS consistently stood out in my work by offering a container ecosystem that balances agility with trust:</p><ul><li><strong>Amazon ECS</strong> → simple, stable orchestration</li><li><strong>Amazon EKS</strong> → Kubernetes when you need the ecosystem</li><li><strong>AWS Fargate</strong> → no servers, no patching, no CVE headaches</li><li><strong>Amazon ECR</strong> → image scanning and secure image delivery</li><li><strong>CloudWatch + OpenTelemetry</strong> → unified observability</li><li><strong>IAM + Secrets Manager</strong> → security built into the platform</li></ul><p>In finance, reliability is non-negotiable.</p><p>AWS gives you the building blocks to innovate without breaking that trust.</p><h3>2. Security-first design: why AWS fits regulated industries</h3><p>Coming from a cybersecurity background (CISSP), I’ve always followed one rule:</p><p><strong>“Security should be a default, not an afterthought.”</strong></p><p>AWS naturally enforces this mindset:</p><ul><li>IAM roles for ECS tasks and EKS pods (IRSA)</li><li>ECR scan-on-push</li><li>Private Bedrock and SageMaker endpoints</li><li>Encrypted traffic via KMS</li><li>Strict network boundaries with Security Groups</li><li>Minimized OS exposure through Fargate</li></ul><p>This aligns perfectly with banking expectations:<br> <strong>governance, traceability, encryption, identity-driven access, and predictable operations.</strong></p><p>When mentoring young engineers, I remind them:<br> <em>A good architecture solves problems — a great architecture reduces risk.</em><br> AWS enables that greatness.</p><h3>3. GenAI is exciting — but secure GenAI is the real revolution</h3><p>Organizations no longer ask <em>“Should we adopt GenAI?”</em><br> They ask:</p><ul><li><em>How do we run GenAI securely?</em></li><li><em>How do we keep customer data private?</em></li><li><em>How do we integrate LLMs into existing microservices?</em></li><li><em>What does governance look like for GenAI pipelines?</em></li></ul><p>AWS provides a secure and compliant foundation:</p><ul><li><strong>Amazon Bedrock</strong> → IAM-secured foundation models</li><li><strong>SageMaker</strong> → training, tuning, governance</li><li><strong>VPC endpoints</strong> → no public internet exposure</li><li><strong>KMS</strong> → encrypted inputs and outputs</li><li><strong>CloudTrail</strong> → full auditability</li><li><strong>ECS/EKS</strong> → containerized GenAI microservices</li></ul><p>In many of my projects, containers became the connective tissue:</p><p><strong>API → microservice → Bedrock/SageMaker → business logic</strong></p><p>It’s the perfect balance of innovation and governance.</p><h3>4. The real value is people: mentoring the next generation</h3><p>I’ve built systems for years — but what I enjoy most is <strong>helping people grow</strong>.</p><p>I regularly mentor:</p><ul><li>students preparing AWS certifications</li><li>young engineers entering cloud or cybersecurity</li><li>professionals transitioning into GenAI</li><li>teams learning to deploy secure workloads on AWS</li></ul><p>I share lessons from real incidents, audits, production issues, and successes.</p><p>My philosophy is simple:</p><p><strong>Explain clearly. Guide patiently. Empower others.</strong></p><p>This is why the AWS Community Builder program resonates with me — <br> because these are contributions I already make with passion.</p><h3>5. A practical GenAI + Containers + Security architecture</h3><p>Here’s a clean, high-level blueprint I often recommend.</p><h3>Secure GenAI + Containers Architecture</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TL9fZD3Bqhg7bY0WU4dEnw.png" /><figcaption>Secure GenAI + Containers Architecrure</figcaption></figure><p>This architecture is simple, auditable, and production-ready — ideal for regulated sectors.</p><h3>6. Why I want to contribute more to the AWS community</h3><p>After years of building and securing cloud systems, something became clear:</p><p><strong>Knowledge becomes more valuable when you share it.</strong></p><p>I want to continue:</p><ul><li>simplifying complex cloud topics</li><li>guiding people on their certification journeys</li><li>sharing real-world security and GenAI patterns</li><li>helping professionals transition confidently into cloud roles</li></ul><p>The AWS ecosystem has given me tools, opportunities, and community — <br> now I want to give back even more.</p><h3>Conclusion: technology evolves, but trust remains the foundation</h3><p>Containers changed how we package software.<br> GenAI is changing how we build intelligence.<br> Cloud changes how we scale.</p><p>But trust — in the system, the architecture, and the people — is what makes all of this matter.</p><p>AWS provides the balance of innovation and security that regulated environments need.<br> And communities like AWS Community Builders make this journey meaningful.</p><h3>About the Author</h3><p><strong>Bangaly Kaba, PhD, CISSP</strong> is a Cloud &amp; Security Expert with 15+ years of experience in banking and financial services.<br> He specializes in secure cloud architectures, containerized workloads, and Generative AI in regulated environments.<br> He mentors young professionals in cloud, security, and AWS certifications, sharing practical insights with the global cloud community.<br> His mission is simple: help others build safely, confidently, and intelligently on AWS.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6b9c3f0cab02" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Cloud computing is a vast and dynamic universe, and my journey to mastering Kubernetes — what I…]]></title>
            <link>https://medium.com/@fbkaba/cloud-computing-is-a-vast-and-dynamic-universe-and-my-journey-to-mastering-kubernetes-what-i-184984315646?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/184984315646</guid>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Sun, 12 Oct 2025 19:47:49 GMT</pubDate>
            <atom:updated>2025-10-12T19:47:49.024Z</atom:updated>
            <content:encoded><![CDATA[<p>Cloud computing is a vast and dynamic universe, and my journey to mastering Kubernetes — what I call my Kubestronaut journey — has been nothing short of an exhilarating space mission. From refactoring monolithic banking applications to navigating the complexities of cloud-native architectures, this adventure has shaped my skills, mindset, and approach to modern DevOps practices.</p><h3>Liftoff: The Beginning of My Kubernetes Voyage 🚀</h3><p>It all started when I got the opportunity to work on a credit claim banking application <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Banque de France</a> . The system was a classic monolith, and I was tasked with refactoring it into microservices — a transformation that made containerization a perfect fit.</p><p>At that time, I was introduced to Docker and Kubernetes, and I quickly realized that mastering these tools was crucial to standing out in the cloud ecosystem. But knowing the basics wasn’t enough — I needed to certify my expertise.</p><p>To get started, I joined weekend batch classes with <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Edureka</a> to build a solid foundation in DevOps basics. From there, my day-to-day tasks revolved around:</p><ul><li>Architecting, refactoring and developping</li><li>Writing Dockerfiles and building container images</li><li>Creating Kubernetes manifest files for various objects (Pods, Deployments, Services, ServiceAccounts, Volumes, PersistentVolumeClaims)</li><li>Exploring Helm for package management</li><li>Building containerized CI/CD pipelines</li><li>Deploying in different environments</li></ul><h3>Reaching for the Stars: The Path to Kubernetes Mastery</h3><p>To accelerate my Kubernetes learning and truly become a Kubestronaut, I set a goal to achieve the CKAD (Certified Kubernetes Application Developer), CKA (Certified Kubernetes Administrator), and CKS (Certified Kubernetes Security Specialist) certifications.</p><p>Here’s how I made it happen:</p><h3>1️⃣ The Ultimate Learning Resources</h3><ul><li><a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Mumshad Mannambeth</a> <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">KodeKloud</a> — The best hands-on Kubernetes training</li><li><a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Kim Wüstkamp</a> <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Killer Shell</a> — The ultimate exam simulator for CKA/CKAD/CKS</li></ul><p>These two resources alone can take you from zero to Kubernetes hero in no time!</p><h3>2️⃣ Practice, Practice, Practice!</h3><p>The Kubernetes exams are practical, time-bound, and intense — which means speed and efficiency are key. I focused on:</p><ul><li>Mastering Vim, Kubernetes shortcuts, and aliases to work faster</li><li>Navigating <a href="http://kubernetes.io/">kubernetes.io</a> docs efficiently to grab YAML templates and modify them quickly</li><li>Running countless labs and scenarios to reinforce concepts</li></ul><h3>3️⃣Adapting to New Kubernetes Trends</h3><p>Kubernetes is evolving rapidly, and staying up-to-date is crucial. Recent additions to the exam topics include:</p><ul><li>Kustomize — Managing Kubernetes configurations</li><li>Helm — Simplifying Kubernetes app deployments</li><li>ArgoCD — GitOps-powered continuous delivery</li><li>CRDs (Custom Resource Definitions) — Extending Kubernetes functionality</li><li>Gateway Classes — Modern API gateway management</li></ul><h3>Joining the Kubernetes Community 🌍</h3><p>One of the best ways to grow in Kubernetes is to connect with the right people. I followed top Kubernetes practitioners like <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Cloud Native Computing Foundation (CNCF)</a> <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Linux Foundation Education</a> <a href="https://www.linkedin.com/article/edit/7309632105804681216/#">Kubestronaut</a>, who shared their insights and strategies for mastering Kubernetes. The community’s support and knowledge-sharing played a crucial role in my journey.</p><h3>Final Thoughts: Becoming a Kubestronaut</h3><p>My Kubernetes journey has been like a rocket ride through the cloud — full of challenges, learning, and growth. If you’re looking to become a Kubestronaut, here’s my final advice:</p><p>✅ Get certified (CKAD, CKA, CKS, KCNA, KCSA) to validate your skills</p><p>✅ Follow top instructors (<a href="https://www.linkedin.com/in/mmumshad?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAAAYAKJUBsbYj9F3qIlJAh3Hn59QCX75LlkY">Mumshad Mannambeth</a> , <a href="https://www.linkedin.com/in/kimwuestkamp?miniProfileUrn=urn%3Ali%3Afs_miniProfile%3AACoAAAw-XWABoGnA__TD-pqyT-yFwft--B_-VfI">Kim Wüstkamp</a>) for structured learning</p><p>✅ Practice relentlessly to gain hands-on expertise</p><p>✅ Master speed and efficiency (Vim, CLI, YAML templates)</p><p>✅ Stay updated with new Kubernetes trends (Helm, ArgoCD, CRDs, Gateway Classes…)</p><p>✅ Engage with the Kubernetes community for guidance and motivation</p><p>Kubernetes is not just a tool — it’s a mindset. Embrace it, explore it, and soon, you’ll be navigating the cloud-native universe like a true Kubestronaut! 🚀</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=184984315646" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[️ Supercharging Restaurant Discovery with Agentic AI]]></title>
            <link>https://medium.com/@fbkaba/%EF%B8%8F-supercharging-restaurant-discovery-with-gemini-2-5-mcp-and-open-adk-648b448497d6?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/648b448497d6</guid>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <category><![CDATA[gemini]]></category>
            <category><![CDATA[a2a]]></category>
            <category><![CDATA[google-adk]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Mon, 21 Apr 2025 10:07:26 GMT</pubDate>
            <atom:updated>2026-01-18T23:02:33.412Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*1C5BAbhLpijRJT_99v6rXQ.png" /></figure><h3>From Google Gemini + MCP to Amazon Bedrock on AWS</h3><p><strong>Bangaly Kaba</strong><br>Cloud, Security &amp; Generative AI Architect<br><em>Originally published April 21, 2025 — updated with an AWS Generative AI perspective</em></p><h3>Introduction — From Chatbots to Agentic Systems</h3><p>Large Language Models are no longer just conversational interfaces. They are evolving into <strong>agents</strong>: systems capable of reasoning, acting, and interacting with the real world through tools and APIs.</p><p>This architectural shift — often referred to as <strong>agentic AI</strong> — changes how we design intelligent applications. Instead of relying on static knowledge, agentic systems can now:</p><ul><li>Use external tools and APIs in real time 🌐</li><li>Discover capabilities dynamically 🤖</li><li>Compose actions to complete complex tasks 🛠️</li></ul><p>This article explores that shift through a concrete use case — restaurant discovery — starting from a <strong>working Gemini + MCP + Open ADK implementation</strong>, and extending it with <strong>equivalent, production-grade patterns on AWS using Amazon Bedrock</strong>.</p><p>The objective is not vendor comparison, but <strong>architectural maturity</strong>.</p><h3>The Original Implementation: Gemini + MCP + Open ADK (Working Baseline)</h3><p>The initial experiment demonstrates how <strong>Google Gemini 2.5 Flash Preview</strong> can be transformed into a real-time restaurant recommender agent using:</p><ul><li><strong>Model Context Protocol (MCP)</strong> for structured tool discovery</li><li><strong>Open ADK</strong> as an agent server exposing callable capabilities</li><li><strong>TripAdvisor API (via RapidAPI)</strong> for live restaurant data</li></ul><p>The system is fully functional:</p><ol><li>The model interprets a user goal (e.g. <em>“Find the best restaurants in Paris”</em>)</li><li>Available tools are discovered via MCP metadata</li><li>APIs are invoked dynamically through Open ADK</li><li>The model returns grounded, real-time recommendations</li></ol><p>This implementation remains the <strong>reference working demo</strong> and proof of hands-on agentic orchestration.</p><h3>What MCP Teaches Us (Beyond Any Single Platform)</h3><p>MCP introduces several architectural ideas that generalize well:</p><ul><li>Tool discovery via metadata</li><li>Structured context injection</li><li>Action-to-Action (A2A) chaining</li><li>Observability of agent decisions</li></ul><p>These concepts are <strong>not Gemini-specific</strong>. They translate directly to how agentic systems are designed on AWS.</p><p>Agentic AI is a <strong>design pattern</strong>, not a product feature.</p><h3>AWS Perspective — Agentic AI with Amazon Bedrock</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*0kutT6-He_Wwb11G.png" /><figcaption><strong>Reference architecture for an agentic assistant on AWS using Amazon Bedrock Agents.</strong></figcaption></figure><h3>Agent Core — A Conceptual Decomposition</h3><p>From an architectural perspective, an agent can be viewed as a composition of three logical layers.<br> These are <strong>conceptual building blocks</strong>, not standalone AWS services.</p><h4>1. Agent Runtime</h4><p>The runtime is responsible for:</p><ul><li>Interpreting user intent</li><li>Planning actions</li><li>Coordinating tool execution</li></ul><p>On AWS, this role is fulfilled by <strong>Amazon Bedrock Agents</strong>, which handle reasoning and action planning without requiring infrastructure management.</p><h4>2. Agent Memory</h4><p>Memory enables agents to:</p><ul><li>Maintain conversational context</li><li>Store intermediate results</li><li>Reuse prior knowledge when appropriate</li></ul><p>In AWS architectures, agent memory is typically implemented using managed services such as <strong>Amazon DynamoDB</strong> or <strong>Amazon S3</strong>, combined with prompt-level context injection.</p><h4>3. Agent Gateway</h4><p>The gateway layer exposes external capabilities to the agent in a controlled way:</p><ul><li>APIs</li><li>Databases</li><li>Third-party services</li></ul><p>On AWS, this is implemented using <strong>AWS Lambda</strong> and <strong>Amazon API Gateway</strong>, with explicit schemas that define what the agent is allowed to invoke.</p><p>Together, these layers form an <strong>agent core architecture</strong> that is observable, secure, and evolvable — aligning naturally with AWS Well-Architected principles.</p><h3>Conceptual Mapping</h3><p>MCP / Open ADK conceptAWS equivalentTool discoveryBedrock Agents + tool schemasTool executionAWS LambdaAPI exposureAmazon API GatewayContext managementPrompt templates + agent memoryA2A orchestrationAgent planning &amp; Lambda chaining</p><p>The principle remains the same:</p><p>The model reasons, tools act, and the system stays observable, secure, and controlled.</p><h3>Amazon Bedrock Teaser — Agent-Based Reasoning (Preview)</h3><p>Below is a <strong>minimal, illustrative teaser</strong> showing how the restaurant agent is invoked using <strong>Amazon Bedrock Agents</strong>.<br>It mirrors the Gemini + MCP flow without claiming a full AWS rewrite yet.</p><pre>import boto3<br><br># Requires an existing Bedrock Agent + Alias<br>client = boto3.client(&quot;bedrock-agent-runtime&quot;, region_name=&quot;us-east-1&quot;)<br>response = client.invoke_agent(<br>    agentId=&quot;restaurant-discovery-agent&quot;,<br>    agentAliasId=&quot;prod&quot;,<br>    sessionId=&quot;demo-session-001&quot;,<br>    inputText=&quot;Find the best restaurants in Paris with good reviews and a romantic vibe.&quot;<br>)<br># Stream the agent response<br>for event in response[&quot;completion&quot;]:<br>    if &quot;chunk&quot; in event:<br>        print(event[&quot;chunk&quot;][&quot;bytes&quot;].decode(&quot;utf-8&quot;), end=&quot;&quot;)</pre><p>In this model:</p><ul><li>The <strong>agent plans the actions</strong></li><li><strong>Lambda-backed tools</strong> call external APIs (e.g. TripAdvisor)</li><li>IAM policies enforce access boundaries</li><li>The model synthesizes a grounded response</li></ul><p>This is the AWS-native equivalent of MCP + Open ADK, with built-in governance.</p><h3>Demo &amp; Repositories — Honest Positioning</h3><h3>Existing Repository (Working Demo)</h3><p>👉 <strong>Gemini + MCP + Open ADK baseline implementation</strong><br>https://github.com/fbkaba/gemini-restaurant-mcp-adk</p><p>This repository contains:</p><ul><li>A runnable MCP agent server</li><li>Live TripAdvisor API integration</li><li>Screenshots and demo instructions</li></ul><p>It remains the <strong>hands-on proof of agentic AI orchestration</strong>.</p><h3>AWS Bedrock Repository (Planned / In Progress)</h3><p>The AWS-native version follows the same agentic architecture:</p><pre>aws-bedrock-restaurant-agent/<br>├── agent/<br>│   ├── agent.yaml              # Bedrock agent definition<br>│   ├── prompts/<br>│   │   └── system_prompt.txt<br>├── lambdas/<br>│   ├── search_restaurants/<br>│   │   └── handler.py<br>│   └── filter_results/<br>│       └── handler.py<br>├── api/<br>│   └── openapi.yaml            # Tool schemas<br>├── infrastructure/<br>│   └── cdk_stack.py            # Bedrock + Lambda + API Gateway<br>├── README.md</pre><p>This enables:</p><ul><li>Clear separation of reasoning and execution</li><li>Serverless scalability</li><li>IAM-based security</li><li>Easy evolution of tools and prompts</li></ul><h3>Why This Matters (Beyond Restaurants)</h3><p>The restaurant discovery use case is intentionally simple.<br>The <strong>architecture is not</strong>.</p><p>The same agentic pattern applies to:</p><ul><li>Travel and hospitality assistants</li><li>E-commerce product discovery</li><li>Internal knowledge agents</li><li>FinTech and customer-support workflows</li></ul><p>What matters is not the domain, but the <strong>agentic discipline</strong>.</p><h3>Final Thoughts — Agentic AI as a Design Skill</h3><p>Agentic AI is not a buzzword.<br>It is a <strong>design skill</strong> that combines:</p><ul><li>Reasoning models</li><li>Explicit tools</li><li>Secure orchestration</li><li>Operational governance</li></ul><p>Whether implemented with <strong>Gemini + MCP</strong> or <strong>Amazon Bedrock + AWS serverless services</strong>, the core principle remains:</p><p><strong>Models become valuable when they can act — safely, transparently, and predictably.</strong></p><h3>Resources</h3><ul><li>Gemini + MCP implementation (baseline):<br>👉 <a href="https://github.com/fbkaba/gemini-restaurant-mcp-adk">https://github.com/fbkaba/gemini-restaurant-mcp-adk</a></li></ul><p><em>This article reflects an evolution from experimental agentic systems to AWS-native, production-oriented Generative AI architectures.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=648b448497d6" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ How I Cleared All AWS Certifications Twice Without a Single Failure — And You Can Too! ☁️]]></title>
            <link>https://medium.com/@fbkaba/how-i-cleared-all-aws-certifications-twice-without-a-single-failure-and-you-can-too-%EF%B8%8F-1daa7319192e?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/1daa7319192e</guid>
            <category><![CDATA[aws-golden-jacket]]></category>
            <category><![CDATA[multi-cloud]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[blue-jacket]]></category>
            <category><![CDATA[continuous-learning]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Sat, 19 Apr 2025 15:47:18 GMT</pubDate>
            <atom:updated>2025-04-19T15:47:18.183Z</atom:updated>
            <content:encoded><![CDATA[<h3>🌟 How I Cleared All AWS Certifications Twice Without a Single Failure — And You Can Too! 💪☁️</h3><p>Hello Cloud Enthusiasts! 👋</p><p>I’m excited to share a personal achievement that still makes me smile:<br>I’ve successfully cleared <strong>all AWS certifications not just once, but TWICE — without a single failure.</strong> 🎉</p><p>From <strong>Cloud Practitioner</strong> to the advanced <strong>SAP on AWS</strong>, this journey has been nothing short of transformative.</p><h3>🔥 The Secret Sauce? Passion + Strategy + Practice</h3><p>I’m deeply passionate about learning and always curious about what’s next in cloud. That hunger to grow helped me stay focused and inspired.</p><h3>My go-to preparation plan:</h3><ul><li>📚 Followed <strong>AWS SkillBuilder</strong> training paths</li><li>👨‍💻 Did tons of <strong>hands-on labs and exercises</strong></li><li>🧠 Took multiple practice exams on <strong>Whizlabs</strong>, <strong>Udemy</strong>, and <strong>Tutorials Dojo</strong></li><li>🌀 Used a smart <strong>“loop and flag” strategy</strong> during exams</li><li>😴 Prioritized <strong>rest and mindset</strong> — a fresh brain works wonders</li><li>💪 Stayed <strong>confident and consistent</strong></li></ul><p>💡 Pro Tip: I barely passed my SAP on AWS — just above the required score. But that’s where <em>preparation meets persistence</em>. It was a close call, but a win nonetheless!</p><h3>🌐 My Multi-Cloud Path Doesn’t Stop at AWS…</h3><p>I’ve also cleared:</p><ul><li>✅ <strong>All Google Cloud certifications</strong></li><li>✅ <strong>19 Microsoft certifications</strong>, including the latest <strong>DP-600 &amp; DP-700</strong></li><li>✅ Cloud certs from <strong>CNCF (Kubernetes), Alibaba Cloud, Tencent, Oracle, Nutanix, Aviatrix</strong></li></ul><p>Cloud is evolving fast, and staying <strong>multi-cloud fluent</strong> helps me stay ahead.</p><h3>🚀 Key Takeaways to Boost Your Cloud Journey</h3><ol><li><strong>Get hands-on</strong> — practice labs and real-world scenarios</li><li><strong>Use multiple sources</strong> — combine SkillBuilder, Whizlabs, Udemy, Tutorials Dojo</li><li><strong>Rest and reset</strong> — don’t underestimate a good night’s sleep</li><li><strong>Have a strategy</strong> — loop through, flag questions, and review</li><li><strong>Stay confident</strong> — trust yourself, even when it gets tough</li></ol><h3>🌟 You’re Stronger Than You Think</h3><p>This certification journey made me <strong>more resilient, more skilled, and more determined</strong>. If you’re just beginning or eyeing your next badge — keep going. Your hard work will pay off. 💯</p><p>Let’s grow and inspire each other.</p><p>Below you’ll find my aws exam history:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CAvn1QVsyqBzMwtEklAyew.png" /><figcaption>AWS-exam-history-1</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oAugcIygKqzEGKUPXkaIQg.png" /><figcaption>AWS-exam-history-2</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*5II0Z0nHnazSSEhMrRzn4w.png" /></figure><p>🫶 Connect with me on <a href="https://www.linkedin.com/in/bangaly-kaba-phd-6aa51925/">LinkedIn</a>, <a href="https://x.com/@bkaba_bma">X</a>, <a href="https://github.com/fbkaba">Github</a> and even <a href="https://www.strava.com/athletes/bangalyk">Strava</a> to talk tech — or swap running and biking tips! 🏃‍♂️🚴‍♀️</p><h4>#AWS #CloudCertifications #LearningJourney #MultiCloud #GCP #Azure #DevOps #Kubernetes #Confidence #CertificationTips #CloudLife #GoldenJacket #BlueJacket</h4><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1daa7319192e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Multimodal YouTube Video Summarization]]></title>
            <link>https://medium.com/@fbkaba/using-google-gemini-flash-for-youtube-video-summarization-0e1a828f0244?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/0e1a828f0244</guid>
            <category><![CDATA[gemini]]></category>
            <category><![CDATA[google-gemini-flash]]></category>
            <category><![CDATA[generative-ai-use-cases]]></category>
            <category><![CDATA[google-cloud-platform]]></category>
            <category><![CDATA[google-ai-studio]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Sun, 23 Mar 2025 17:14:03 GMT</pubDate>
            <atom:updated>2026-01-18T20:44:44.395Z</atom:updated>
            <content:encoded><![CDATA[<h3>From Google Gemini Flash Thinking to Amazon Bedrock on AWS</h3><p><strong>Bangaly Kaba</strong><br>Cloud, Security &amp; Generative AI Architect</p><h3>Introduction</h3><p>Multimodal Generative AI is rapidly changing how we extract value from rich media such as videos, conferences, and online content. Early experiments with models like <strong>Google Gemini Flash Thinking</strong> demonstrate how text and video can be analyzed together to generate concise summaries.</p><p>However, moving from experimentation to <strong>production-ready systems</strong> requires more than a single multimodal API call.</p><p>In this article, I revisit a YouTube video summarization use case initially explored with Google Gemini Flash and show how the <strong>same reasoning patterns can be operationalized on AWS using Amazon Bedrock</strong>, following cloud-native and Well-Architected principles.</p><h3>Use Case Overview</h3><p>The objective is to analyze a YouTube video in order to:</p><ul><li>Identify active speakers</li><li>Summarize their participation</li><li>Produce a structured, reusable summary</li></ul><p>Typical applications include:</p><ul><li>Technical talks and conference recordings</li><li>Meetups and training sessions</li><li>Internal knowledge-sharing platforms</li></ul><h3>Baseline Prototype: Google Gemini Flash Thinking</h3><p>As a first prototype, Google Gemini Flash Thinking can reason directly over multimodal inputs (video + text prompts).</p><h3>Input Components</h3><ul><li><strong>Prompt text:</strong><br>“Identify active speakers and more details about their attendance”</li><li><strong>Prompt video:</strong><br>YouTube video URL</li><li><strong>System instruction:</strong><br>“Behave as a content analyzer for summarization”</li></ul><h3>Sample Python Implementation (Prototype)</h3><pre>from google import genai<br>from google.genai import types<br><br>GEMINI_API_KEY = &quot;xxxxxxxxxxxxx&quot;<br>def generate(video_url, prompt):<br>    client = genai.Client(api_key=GEMINI_API_KEY)<br>    model = &quot;gemini-2.0-flash-thinking-exp-01-21&quot;<br>    contents = [<br>        types.Content(<br>            role=&quot;user&quot;,<br>            parts=[<br>                types.Part.from_uri(<br>                    file_uri=video_url,<br>                    mime_type=&quot;video/*&quot;,<br>                ),<br>                types.Part.from_text(<br>                    text=&quot;Identify active speakers and more details about their attendance&quot;<br>                ),<br>            ],<br>        ),<br>    ]<br>    config = types.GenerateContentConfig(<br>        temperature=0.7,<br>        max_output_tokens=4096,<br>        system_instruction=[types.Part.from_text(text=prompt)],<br>    )<br>    with open(&quot;video-summary.txt&quot;, &quot;w&quot;) as file:<br>        for chunk in client.models.generate_content_stream(<br>            model=model,<br>            contents=contents,<br>            config=config,<br>        ):<br>            print(chunk.text, end=&quot;&quot;)<br>            file.write(chunk.text)<br>if __name__ == &quot;__main__&quot;:<br>    generate(<br>        &quot;https://www.youtube.com/watch?v=YLEUVtuh0jM&quot;,<br>        &quot;Identify active speakers and more details about their attendance&quot;<br>    )</pre><p>This prototype highlights the <strong>power of multimodal reasoning</strong>, but it also reveals limitations when considering scalability, security, and governance.</p><h3>AWS-Native Perspective: From Prototype to Production</h3><p>In real-world environments, video summarization is best implemented as a <strong>pipeline</strong>, not a single model invocation.</p><p>On AWS, this use case naturally maps to a cloud-native architecture using:</p><ul><li>Amazon S3 for video and audio storage</li><li>Amazon Transcribe for speaker-aware transcription</li><li><strong>Amazon Bedrock</strong> for reasoning and summarization</li><li>AWS Lambda for serverless orchestration</li></ul><p>This approach enables:</p><ul><li>Clear separation of concerns</li><li>IAM-based security and data isolation</li><li>Cost-aware, scalable execution</li></ul><p>Rather than embedding business logic inside one model call, <strong>Amazon Bedrock allows Generative AI to be operationalized responsibly</strong>, with model choice flexibility (Claude, Titan, Llama) and no infrastructure management.</p><h3>Amazon Bedrock Sample: Speaker-Aware Summarization (Preview)</h3><p>On AWS, the reasoning step is handled by <strong>Amazon Bedrock</strong>, which allows invoking foundation models without managing infrastructure.</p><p>Below is a <strong>minimal example</strong> showing how a speaker-attributed transcript can be summarized using Amazon Bedrock.</p><pre>import boto3<br>import json<br><br>bedrock = boto3.client(&quot;bedrock-runtime&quot;, region_name=&quot;us-east-1&quot;)<br>prompt = &quot;&quot;&quot;<br>You are an AI content analyzer.<br>Given the transcript below with speaker labels,<br>identify active speakers and summarize their participation.<br>Transcript:<br>[SPEAKER_0]: Welcome everyone to today&#39;s session.<br>[SPEAKER_1]: I will present the architecture and key design choices.<br>&quot;&quot;&quot;<br>body = {<br>    &quot;anthropic_version&quot;: &quot;bedrock-2023-05-31&quot;,<br>    &quot;messages&quot;: [<br>        {<br>            &quot;role&quot;: &quot;user&quot;,<br>            &quot;content&quot;: prompt<br>        }<br>    ],<br>    &quot;max_tokens&quot;: 600,<br>    &quot;temperature&quot;: 0.3<br>}<br>response = bedrock.invoke_model(<br>    modelId=&quot;anthropic.claude-3-sonnet-20240229-v1:0&quot;,<br>    body=json.dumps(body)<br>)<br>result = json.loads(response[&quot;body&quot;].read())<br>print(result[&quot;content&quot;][0][&quot;text&quot;])</pre><h3>High-Level AWS Architecture</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*01ZCOI-PkB9EpJy9.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/824/0*Fdz9I26QajSpt7BG.jpg" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/900/0*0UbkfWVDCyFjpKa-.jpg" /></figure><p>This use case follows a <strong>pipeline-based approach</strong>:</p><ol><li>Video or audio stored in Amazon S3</li><li>Speaker diarization performed by Amazon Transcribe</li><li>Transcript analyzed by Amazon Bedrock</li><li>Serverless orchestration with AWS Lambda</li></ol><p>This architecture aligns with <strong>AWS Well-Architected principles</strong> and enables security, scalability, and cost control.</p><h3>Understanding the End-to-End Workflow</h3><ul><li>Media content is stored in Amazon S3</li><li>Amazon Transcribe produces a transcript with speaker labels</li><li>Amazon Bedrock analyzes the transcript to:</li><li>Identify active speakers</li><li>Summarize contributions</li><li>Generate structured outputs</li><li>Results are stored for reuse, search, or analytics</li></ul><p>This pattern is suitable for both <strong>community-scale</strong> and <strong>enterprise-grade</strong> workloads.</p><h3>Conclusion</h3><p>Multimodal video summarization is not just about passing a video to a foundation model. Production systems require <strong>governance, security, cost control, and observability</strong>.</p><p>While Google Gemini Flash Thinking provides a valuable baseline for multimodal experimentation, <strong>Amazon Bedrock enables production-grade Generative AI on AWS</strong>, making it possible to build secure, scalable, and reusable video analysis pipelines.</p><p>This article reflects an evolution from experimentation to <strong>operational Generative AI</strong> — a key challenge many builders face today.</p><h3>Resources</h3><p>Prototype walkthrough (Gemini baseline):<br>https://github.com/fbkaba/gemini-flash-thinking-walkthough.git</p><p><em>This article serves as a bridge toward a deeper, AWS-native series focused on Amazon Bedrock, including detailed architecture, prompt design, and production best practices.</em></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=0e1a828f0244" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Journey of GenAI with AWS Bedrock based sample Images]]></title>
            <link>https://medium.com/@fbkaba/a-journey-of-genai-with-aws-bedrock-based-sample-images-65a9fc0ef05f?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/65a9fc0ef05f</guid>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Tue, 31 Dec 2024 13:37:02 GMT</pubDate>
            <atom:updated>2024-12-31T13:37:02.978Z</atom:updated>
            <content:encoded><![CDATA[<p>In the not-so-distant past, machines were relegated to mere calculation and automation. Today machines can create, innovate, and even rival human imagination! Is GenAI slowly dissolving the lines between human creativity and artificial intelligence? Let’s journey through the early days of Generative AI…</p><p><strong><em>Before that, here is a little back story of “how I met Bangaly”</em></strong></p><p>I have known and admired Bangaly’s professional accomplishments for more than a year now. He has over 1000 Badges, a CNCF Kubestronaut, a AWS Golden Jacket , 17xAWS, 14xGCP, 17xMicrosoft, 6xComptia, 4xHashicorp, 4xGithub, 3xOCI, and many more. I couldn’t comprehend how he achieved that, so I did the only logical thing, <strong><em>ask him</em></strong>!</p><p>The answer wasn’t anything we all haven’t heard before. There was the irritating <strong><em>consistency</em></strong>, the annoying <strong><em>hardwork</em></strong>, never missing <strong><em>curiosity</em></strong> and a the magical <strong><em>kindness</em></strong></p><p>View the crazy things he has accomplished here ☞<a href="https://www.linkedin.com/in/bangaly-kaba-phd-6aa51925/">Bangaly’s LinkedIn</a></p><p>We are kind of an odd duo, with a shared passion to encourage, empower and share knowledge with our community. We shall be creating series of content surrounding the realm of AI, ML and all the fun stuff around Cloud Computing.</p><p>Let’s start from the beginning…..</p><p><strong><em>Artificial Intelligence has been lingering around since 1941!</em></strong></p><p>In 1950, <strong>Alan Turing</strong> published a paper titled <em>“Computing Machinery and Intelligence”</em> in which he proposed the imitation game.<br>The game involves: Three participants: <em>a human interrogator, a human respondent, and a machine</em>.</p><p>The interrogator’s goal is to determine which of the other two participants is the human and which is the machine.<br>Communication is limited to text-only exchanges.</p><p>The machine aims to fool the interrogator into thinking it is human. Turing argued that if the machine could consistently fool human interrogators, it should be considered to exhibit intelligent behavior. The term Artificial Intelligence was officially coined on august 31, 1955. <a href="http://jmc.stanford.edu/articles/dartmouth.html">Here is the link to the article</a></p><p><strong><em>What then is machine learning?</em></strong></p><p>Machine Learning is a subset of Artificial intelligence. The idea behind machine learning is to feed large amounts of data into algorithms, which can then identify patterns and relationships in the data, and use that knowledge to make predictions or decisions without relying on hard-coded rules. Arthur Samuel, an IBM researcher, is credited with coining the term “Machine Learning” in 1959.</p><p><strong><em>and Deep Learning is…</em></strong></p><p>Deep Learning is a specific type of machine learning that gained widespread attention in the 21st century. It is inspired by the structure and function of the human brain and involves the use of <strong><em>artificial neural networks (ANNs)</em></strong>, which are computational models that mimic the interconnected neurons in the brain.</p><p>Deep learning algorithms can automatically learn complex patterns and representations from raw data, making them highly effective for tasks like image recognition, natural language processing, and speech recognition. While the fundamental concepts of neural networks date back to the 1940s and 1950s, with pioneers like Warren McCulloch and Walter Pitts, deep learning only became practically viable in the late 2000s due to advances in computing power, availability of large datasets, and algorithmic improvements.</p><p><strong><em>In summary,</em></strong></p><p><strong><em>AI</em></strong> is the overarching field, <strong><em>machine learning</em></strong> is a subset of AI that focuses on learning from data, and <strong><em>deep learning</em></strong> is a specific type of machine learning that uses artificial neural networks.</p><p>These fields have evolved over time, with each advancement building upon the foundations laid by previous researchers and breakthroughs.</p><p><strong><em>A Question we should ask ourselves is, if AI has been around for this long, why is it gaining popularity decades after?</em></strong></p><p>Early AI researchers faced several significant challenges as they worked to develop artificial intelligence in the 1950s and 1960s. These included:</p><p><em>Limited Computing Power:</em> Early computers lacked the processing power and memory needed for complex AI algorithms and large data processing.</p><p><em>Unrealistic Expectations:</em> Early AI pioneers were overly optimistic about achieving human-level AI quickly, leading to disappointment when progress was slower.</p><p><em>Lack of Understanding of Intelligence:</em> Researchers underestimated the complexity of human cognition, thinking it could be easily replicated in machines.</p><p><em>Narrow Focus:</em> Early AI research focused on specific tasks like chess, which didn’t translate well to general intelligence or common sense reasoning.</p><p><em>Funding Challenges:</em> AI research experienced cycles of enthusiasm and funding, followed by “AI winters” where interest and financial support dried up.</p><p><em>Software Limitations:</em> The bottleneck in AI development was often software, as creating programs that mimicked human reasoning was very difficult.</p><p><em>Lack of Data:</em> Early researchers didn’t have access to the large datasets needed for effective AI training, limiting their models’ effectiveness.</p><p><strong><em>How have we overcomed these challenges?</em></strong></p><p>Access to Artificial Intelligence has been democratized and its capabilities are continuously advancing. It has also led to generative AI, a game-changing technology.</p><p><strong><em>Generative AI</em></strong> refers to artificial intelligence models that can generate new data, such as texts, images, audios, or videos, based on the patterns and relationships learned from training data, and it is still evolving.</p><p>One of the key players in democratizing access to generative AI is <strong>AWS Bedrock</strong>, a comprehensive platform developed by Amazon Web Services (AWS) that provides developers and researchers with the tools, infrastructure, and resources needed to build and deploy generative AI models.</p><p>AWS Bedrock leverages the power of cloud computing, offering scalable and cost-effective solutions that enable users to train and run large-scale generative AI models without the need for expensive on-premises hardware. By removing barriers to entry, AWS Bedrock has empowered individuals and organizations of all sizes to explore and leverage the capabilities of generative AI.</p><p><strong><em>AWS Bedrock use-case.</em></strong></p><p>Let’s illustrate one of the capability of AWS Bedrock using AWS management console.</p><p><strong><em>How to create an Image on AWS Bedrock</em></strong><br>This can be done in 3 simple steps</p><p><strong><em>Step 1</em></strong></p><p>You require an AWS Account, which is free. Be aware that most of the models are third party tools and will be charged separately, even though you have AWS Credits. We learned the hard way 🙃💸<br>Log into the account and move to the AWS Bedrock UI.</p><p>AWS Bedrock UI and its functionalities</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*NIY0fGJVlOTkIZln1ewSmQ.png" /></figure><p><strong><em>Step 2</em></strong></p><p>Request for access from the desired model</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*qwdbN0hCGIivC-pUob4GDA.png" /></figure><p><strong><em>Step 3</em></strong></p><p>Image generated from desired image</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*ZkaDaXMZ42_99G2omN2gNw.png" /></figure><p><a href="https://github.com/aws-samples/amazon-bedrock-samples">AWS Samples</a> contains pre-built examples to help customers get started with the Amazon Bedrock service.</p><p>We shall be exploring more of them in our next content.</p><p>Stay tuned….</p><p><strong><em>Bangaly and Endah</em></strong></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65a9fc0ef05f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Navigating the New AWS AI Certifications:
A Conversation Between Enthusiasts]]></title>
            <link>https://medium.com/@fbkaba/navigating-the-new-aws-ai-certifications-a-conversation-between-enthusiasts-9e1b48f43a37?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/9e1b48f43a37</guid>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Mon, 30 Dec 2024 10:00:23 GMT</pubDate>
            <atom:updated>2024-12-30T10:00:23.575Z</atom:updated>
            <content:encoded><![CDATA[<p>After recently posting about our experiences with the new AWS AI Practitioner (AIF-C01) and AWS Machine Learning Engineer Associate (MLA-C01) certifications, we were overwhelmed by the number of enthusiastic people who reached out with questions. In response to this incredible interest, we decided to write this article to share our insights with a broader audience of cloud and AI enthusiasts.</p><h3>Who are we?</h3><p>We’re Bangaly Kaba, PhD, and Miklós Csécsi, two cloud enthusiasts at different stages of our AWS and cloud journey. We recently sat down to chat about our experiences with the new AWS AI Practitioner (AIF-C01) and AWS Machine Learning Engineer Associate (MLA-C01) exams.</p><p><strong>Bangaly</strong> is a Senior Tech Lead &amp; Cloud Architect with over 20 years of experience, including a role as a CISO. He holds a PhD in Computer Science, specializing in data mining and natural language processing, as well as a TOGAF certification in enterprise architecture. Bangaly’s expertise spans cloud architecture, blockchain, data analysis, and machine learning. He’s fully certified in multiple cloud platforms and Kubernetes and has a deep background in AI and machine learning. In his current role at Société Générale, he designs and implements innovative solutions for the banking sector. Bangaly’s impressive career includes work on semantic data processing, encryption and decryption modules, secure exchange connectors, single sign-on authentication, and anti-ransomware defense strategies. He’s also experienced in designing payment and messaging applications for numerous operators across multiple countries.</p><p><strong>Miklós</strong> comes from a High-Voltage Electrical Engineering background and has a diverse career in technology. He worked as a Senior Software Developer at Siemens, working on safety- and security-relevant software.Later, he transitioned to a Senior Security Solutions Engineer role, specializing in Application and Cloud Security, working for companies like Snyk and Rapid7. <br>Throughout his career, Miklós has had the opportunity to collaborate with over 250 companies across more than 15 countries, focusing on Cloud- and Application Security and Vulnerability Management. With around five years of hands-on cloud experience and multiple AWS certifications, Miklós is now preparing to start a new chapter as a Solutions Architect at AWS. He has also had the privilege of speaking at several events, from Hamburg through Bern to Cape Town, where he shared insights on cloud and network security. Additionally, he regularly holds workshops and enablement sessions on cloud and IT Security.</p><p>Miklós: Bangaly, it’s great to catch up! I know you’ve recently passed both the AI Practitioner and ML Engineer Associate exams. What was your motivation for taking these?</p><p>Bangaly: Well, Miklós, I took these exams to expand my knowledge and skills in AI and ML, as they’re rapidly advancing technologies with numerous applications across industries. Additionally, having these certifications can enhance my career prospects and open up new opportunities.</p><p>Miklós: I completely agree. I’m convinced that AI will be as transformative as the internet was in its early days. What kind of experience did you have with AI before pursuing these exams?</p><p>Bangaly: I share your enthusiasm for the transformative potential of AI! Before taking these exams, I had some exposure to AI through online courses and tutorials but wanted a more comprehensive understanding of the field. The exams provided me with a structured learning path and the opportunity to apply my knowledge in real-world scenarios.</p><p>Miklós: My background is a bit different. I studied High-Voltage Electrical Engineering, which included coursework in Advanced Mathematics, Stochastics, and Programming. But I haven’t applied AI concepts directly in my professional work yet. When did you first encounter generative AI?</p><p>Bangaly: Your background in Electrical Engineering and your knowledge of Advanced Mathematics and Programming will definitely come in handy for your AI journey.</p><p>I first encountered generative AI at AWS re:Invent 2023 and later at Google Cloud Next ’24, where I was introduced to generative adversarial networks (GANs) for image generation. It was amazing to see how these networks could create realistic images from random noise. Since then, generative AI has evolved a lot and is now widely used in things like image, speech, and text generation.</p><p>Miklós: For me, it was about four months ago at the AWS Summit in Berlin. Viktoria Semaan’s session opened my eyes to the possibilities of generative AI. Have you been using AI at work?</p><p>Bangaly: That’s great to hear! Viktoria Semaan’s session must have been quite inspiring!</p><p>I’ve been using Copilot in my work for about a year now. It’s a powerful AI tool that assists with coding, provides suggestions, and helps streamline workflows. I’m aiming to work on a project involving natural language processing (NLP) and computer vision (CV) to analyze large amounts of text and images. AI can be instrumental in helping us extract valuable insights from that data.</p><p>Miklós: That’s really interesting. To be honest, I haven’t used much of my university knowledge since graduating 6–8 years ago, but lately, I’ve started using AI tools like ChatGPT to make my daily work easier. It’s really like having an assistant for tasks like drafting emails, brainstorming ideas, or getting quick answers, which frees me up to focus on more strategic work.</p><p>How did you find the new AWS exams?</p><p>Bangaly: It’s wonderful that you’re incorporating AI tools into your work — they can certainly enhance productivity and efficiency!</p><p>Regarding the exams, I found them to be challenging but rewarding. The exams covered a wide range of topics, and I had to study and practice extensively to ensure that I was well-prepared. The exams also required me to apply my knowledge in real-world scenarios, which was a great way to solidify my understanding of the concepts.</p><p>Miklós: I completely agree, and congratulations on passing! I’m curious about your experience with the ML Engineer Associate exam. I have it scheduled in three weeks and am preparing with the official AWS courses and some Udemy ones. How did you find the difficulty level, and which study materials aligned best with the actual exam?</p><p>Bangaly: Thank you, Miklós! The exam was challenging but manageable with proper preparation. A solid understanding of fundamental concepts and AWS services is crucial. I found the official AWS courses helpful since they provided a structured learning path. Everyone’s study habits differ, but I recommend finding what works best for you.</p><p>Miklós: That’s really helpful, thank you. Do you have any last-minute tips for someone about to take the exam?</p><p>Bangaly: Well, as you approach the exam, it’s important to remember to take your time and carefully read the questions. Some questions may require good analysis or apply your knowledge in a specific context, so it’s important to understand the question and the scenario before answering.</p><p>Review the Exam Prep Plan course on AWS Machine Learning — Associate (MLA-C01).</p><p>Read the official exam guide carefully. Understand the domains, tasks and In-scope Aws services and features. For services involved in ML lifecycle, have a look in corresponding aws faq docs.</p><p>Additionally, it’s a good idea to take practice exams to get a feel for the exam format and the types of questions that are asked. This will help you identify areas where you need to focus your study efforts.</p><p>Finally, don’t forget to take care of yourself during the exam. Make sure to get enough sleep the night before, eat a healthy meal, and stay hydrated. Taking care of yourself can help you perform at your best on the exam.</p><p>Miklós: I appreciate the insights, Bangaly. I’m curious, what topics did you find most prominent in the new AI Practitioner exam? I’d love to compare notes.</p><p>Bangaly: For the AI Practitioner exam, the topics I found most prominent were related to large language models, prompt tuning, fine-tuning, RAG (retrieval-augmented generation), responsible AI, as well as model-related aspects like cost, accuracy, security, and performance. Services like Bedrock, SageMaker, and AWS ML managed services also played a key role.</p><p>How did you find the AI Practitioner exam?</p><p>Miklós: That’s interesting. I found a large portion of the AI Practitioner exam focused on Bedrock and SageMaker, too.</p><p>I found AWS Skillbuilder to be a great resource, especially their Enhanced Exam Prep Plan, which includes about 15 hours of material (with a subscription) and a free set of 20 official practice questions that mirror the exam format. The plan finishes with a 60–70 question quiz that’s identical to the official exam in style and comes with detailed explanations, which I found really helpful.</p><p>I can also recommend Stéphane Maarek’s Udemy course that covered the exam topics well, and also the Cloud Quest Series on AWS Skillbuilder were a great resource.</p><p>I also have to add: the exam has 85 questions (longer than the usual 65), so everyone should prepare for an extended session!</p><p>Did you encounter questions on prompt engineering or measuring the performance of Foundation Models?</p><p>Bangaly: Yes, I did! Prompt engineering came up in both exams: the AI Practitioner exam had a few questions on prompt engineering, while the ML Engineer Associate exam had a few questions on measuring the performance of Foundation Models.</p><p>Miklós: That’s very insightful. Prompt engineering really shows how we can guide AI to produce more accurate outputs. With all the advancements, we’ll also need to watch out for risks like hallucinations or data leaks.</p><p>How do you see yourself applying this AI and ML knowledge in your current role?</p><p>Bangaly: That’s a great point! The potential risks and new challenges associated with generative AI are indeed significant, and it’s important for ML engineers and security pros to stay vigilant and address these issues as they arise.</p><p>Regarding how I see myself applying this AI and ML knowledge in my current role, I work on a team that develops and maintains a large-scale machine learning platform. This platform is used by many teams across our organization, and my job is to ensure that it is reliable, scalable, and efficient. I use AI and ML techniques to improve the performance of the platform, automate certain tasks, and provide insights to our users.</p><p>In terms of how I think this knowledge will shape my future career path, I see myself continuing to work on projects that involve AI and ML. I am particularly interested in natural language processing and computer vision, and I would like to explore these areas further. I also see myself taking on more leadership roles, where I can help guide the development of AI and ML technologies within our organization.</p><p>Miklós: That’s fascinating. In my new role as an AWS Solutions Architect starting in October, I’m excited to help companies leverage AI and ML. I think there’s huge potential to optimize processes and create innovative products across industries.</p><p>Do you see any particular industries where AI and ML will have the biggest impact?</p><p>Bangaly: That’s great to hear! As an AWS Solutions Architect, you will have the opportunity to help companies leverage AI and ML technologies in a variety of ways. Some of the industries that I see having the biggest impact in the near future include healthcare, finance, and retail.</p><p>In healthcare, AI and ML can be used to improve patient outcomes, reduce healthcare costs, and accelerate drug discovery. For example, AI can be used to analyze medical images, predict patient outcomes, and develop personalized treatment plans. In finance, AI and ML can be used to detect fraud, optimize investment portfolios, and improve risk management. For example, AI can be used to analyze financial data and identify patterns that may indicate fraudulent activity. In retail, AI and ML can be used to improve customer experience, optimize supply chain management, and develop personalized marketing campaigns. For example, AI can be used to analyze customer data and recommend products that are likely to be of interest to them.</p><p>Miklós: Those are some really interesting perspectives! I agree that the potential is enormous. It’s both exciting and a bit daunting to think about how quickly the field is evolving.</p><p>Do you think we’ll need to be constantly upskilling to keep up with the pace of AI advancements?</p><p>Bangaly: Yes, the pace of AI advancements is indeed rapid, and it’s important for professionals to be constantly upskilling to stay relevant. This can involve taking courses, attending conferences, and reading research papers. One thing that I find particularly interesting is the intersection of AI and other fields, such as robotics, natural language processing, and computer vision. These fields are rapidly evolving, and it’s important for professionals to be able to work across these boundaries.</p><p>How do you plan to keep up?</p><p>Miklós: I suppose that’s why certifications are so valuable — they keep our skills sharp. I plan to continue my learning path with the ML Engineer Associate exam, and after that, I’m considering the AWS ML Specialty exam to dive deeper into more advanced topics.</p><p>In addition to certifications, I’m planning to get more hands-on experience through personal projects and by applying ML in real-world scenarios. Working on practical use cases will help me stay on top of the latest trends and give me the chance to explore new technologies like computer vision and NLP.</p><p>I have to say, it’s been great discussing this with you, Bangaly, I feel even more motivated now to delve deeper into AI and ML. Thanks for sharing your experiences!</p><p>Bangaly: It’s been my pleasure, Miklós. Best of luck with your ML Engineer Associate exam — I’m sure you’ll do great!</p><p>And to all the learners out there preparing for AWS AI certifications — whether it’s the AI Practitioner, ML Engineer Associate, or any other exam — we wish you the very best of luck! Remember, the journey of learning is just as important as the destination. Stay curious, keep practicing, and don’t be afraid to dive deep into these fascinating technologies.</p><p>Miklós: Thank you! And congratulations again on your certifications. Your insights have been invaluable!</p><p>As a closing note: I couldn’t agree more, Bangaly! Good luck, everyone! The world of AI and Machine Learning is exciting and full of opportunities. These certifications are a great step towards being part of that future. Study hard, but also enjoy the learning process. You’ve got this!</p><h3>Recommended study materials</h3><p>Based on our experience, here are some study materials we found really helpful for preparing for the AWS AI certifications. Keep in mind, this is just our perspective — there are many other great resources out there!</p><h3>AWS AI Practitioner (AIF-C01):</h3><ul><li>Stéphane Maarek:<a href="https://www.udemy.com/course/aws-ai-practitioner-certified/"> Ultimate AWS Certified AI Practitioner AIF-C01</a></li><li>AWS SkillBuilder official resources:</li><li>Enhanced Exam Prep Plan: AWS Certified AI Practitioner (AIF-C01)<br><strong>(</strong><a href="https://explore.skillbuilder.aws/learn/lp/2194/Enhanced%2520Exam%2520Prep%2520Plan%253A%2520AWS%2520Certified%2520AI%2520Practitioner%2520%28AIF-C01%29"><strong>https://explore.skillbuilder.aws/learn/lp/2194/Enhanced%2520Exam%2520Prep%2520Plan%253A%2520AWS%2520Certified%2520AI%2520Practitioner%2520%28AIF-C01%29</strong></a><strong>)</strong></li><li>Exam Prep Official Practice Question Set: AWS Certified AI Practitioner (AIF-C01 — English)<br><strong>(</strong><a href="https://explore.skillbuilder.aws/learn/course/19790/Exam%2520Prep%2520Official%2520Practice%2520Question%2520Set%253A%2520AWS%2520Certified%2520AI%2520Practitioner%2520%28AIF-C01%2520-%2520English%29"><strong>https://explore.skillbuilder.aws/learn/course/19790/Exam%2520Prep%2520Official%2520Practice%2520Question%2520Set%253A%2520AWS%2520Certified%2520AI%2520Practitioner%2520%28AIF-C01%2520-%2520English%29</strong></a><strong>)</strong></li><li>AWS Summit Berlin 2024 — Building an AWS solutions architect agent with Amazon Bedrock (BOA305) <strong>(</strong><a href="https://www.youtube.com/watch?v=8ftKjZyaqNk"><strong>https://www.youtube.com/watch?v=8ftKjZyaqNk</strong></a><strong>)</strong></li><li>Google Cloud CloudSkillBoost: Beginner: Introduction to Generative AI Learning Path<strong> </strong>(<a href="https://www.cloudskillsboost.google/paths/118">https://www.cloudskillsboost.google/paths/118</a>)</li><li>Frank Kane:<a href="https://www.udemy.com/course/aws-certified-ai-practitioner-aif-c01-hands-on-in-depth/"> Mastering AWS Certified AI Practitioner AIF-C01 — Hands On!</a></li><li>Tiny Technical Tutorials (<a href="https://www.youtube.com/@TinyTechnicalTutorials">https://www.youtube.com/@TinyTechnicalTutorials</a>)</li></ul><h3>AWS Machine Learning Engineer (MLA-C01):</h3><ul><li>Stéphane Maarek &amp; Frank Kane: <a href="https://www.udemy.com/course/aws-certified-machine-learning-engineer-associate-mla-c01/">AWS Certified Machine Learning Engineer Associate</a></li><li>AWS SkillBuilder: Enhanced Exam Prep Plan: AWS Machine Learning — Associate (MLA-C01) <a href="https://explore.skillbuilder.aws/learn/public/learning_plan/view/2192/plan">https://explore.skillbuilder.aws/learn/public/learning_plan/view/2192/plan</a></li><li>AWS SkillBuilder: Exam Prep Enhanced Course: AWS Certified Machine Learning Engineer — Associate (MLA-C01 — English) <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/19876/exam-prep-enhanced-course-aws-certified-machine-learning-engineer-associate-mla-c01-english">https://explore.skillbuilder.aws/learn/course/external/view/elearning/19876/exam-prep-enhanced-course-aws-certified-machine-learning-engineer-associate-mla-c01-english</a></li><li>AWS SkillBuilder: Exam Prep Official Pretest: AWS Certified Machine Learning Engineer — Associate (MLA-C01 — English)</li></ul><h3>Conclusion</h3><p>As cloud enthusiasts at different points in our journeys, we both see a lot of value in continuous learning, especially with how fast AI is evolving. Whether you’re a seasoned pro like Bangaly or newer to cloud AI like Miklós, these AWS certifications are a great way to build your knowledge and stay ahead.</p><p>But it’s not just about passing the exam — it’s about really understanding how to use these technologies to solve real-world problems. As AI keeps changing the game across industries, having this knowledge will be a huge asset for anyone in tech.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=9e1b48f43a37" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How I Achieved the Path Towards 12 AWS Certifications Leading to the Golden Jacket]]></title>
            <link>https://medium.com/@fbkaba/how-i-achieved-the-path-towards-12-aws-certifications-leading-to-the-golden-jacket-ea0d0c653d65?source=rss-c5c68eb01121------2</link>
            <guid isPermaLink="false">https://medium.com/p/ea0d0c653d65</guid>
            <category><![CDATA[exam-preparation]]></category>
            <category><![CDATA[aws-certification-journey]]></category>
            <category><![CDATA[aws-golden-jacket]]></category>
            <category><![CDATA[aws]]></category>
            <category><![CDATA[pointers-to-resources]]></category>
            <dc:creator><![CDATA[Bangaly Kaba]]></dc:creator>
            <pubDate>Thu, 25 Jan 2024 22:47:04 GMT</pubDate>
            <atom:updated>2024-01-27T09:43:23.346Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*WwskF6fxUCBmdsVDxkCFNQ.png" /></figure><h4>A Bit About Myself</h4><p>The University of Clermont Ferrand awarded me a <strong>PhD</strong> in Natural Language Processing using graph theory algorithms applied in text and data mining.</p><p>Since my childhood, I have traveled, studied, worked, and visited several countries: Guinea, Algeria, France, England, Germany, Luxembourg, USA, Canada, Mexico, Cyprus, Switzerland, Spain, Italy, Portugal, Netherlands, Belgium, Morocco, Senegal, Mali, and others.</p><p>Before joining<strong> Société Générale</strong>, I worked for several companies including Thalès, EDF, Gemalto, <strong>Banque de France</strong>, and <strong>Onepoint </strong>as a consultant.</p><p>My role at Onepoint was to mentor and deliver AWS courses on various topics such as architecture, development, and operations (sysops).</p><h3>Start of Journey</h3><p>I started my AWS certification journey in<strong> September 2020</strong>. I was inspired by the power of the cloud. Some features I liked straight away were concepts of managed service, serverless, elasticity, pay on demand, and total cost ownership reduction.</p><p>One can focus on their own business and get rid of undifferentiated heavy lifting tasks.</p><p>For the little story, when the COVID-19 epidemic broke out, we were confined. I chatted from time to time with a colleague. He had just started TOGAF online training because he was considering a career as an architect. I also became interested in following the same path. I chose to follow AWS architecting and DevOps courses.</p><p>My choice was motivated by the fact that at this period, I was working on a project to modernize a monolithic banking application. In this context, I developed the application in microservices with Springboot, implemented the Docker files, created the images, and started deploying the Kubernetes manifests such as pods, deployments, services, ingress resources, persistence volume claims, and volumes in the cluster.</p><p>Browsing the web, I saw an online training platform and registered. I took some batch classes on weekends and nights. I followed courses on foundational services: IAM, S3 storage, EC2, RDS, Lambda, CloudWatch, CodeCommit, CodeBuild, CodeDeploy, and CodePipeline. These lessons were accompanied by hands-on exercises and challenges that I completed with my free-tier account. At the end of the first two months, I had my certificates and a solid basis to begin my AWS certification journey.</p><p>Then I went to the AWS certification and training official site. I discovered certification badges along with guidelines and pointers to resources for preparing for them.</p><p>Finally, my cloud journey started. It involved hard work, but believe me, if you focus and spend the effort, it will pay off not only for your dream job but also for all the solutions to business needs that come to you with various cloud services that you couldn’t afford easily with an on-premise datacenter.</p><p>My goal was to understand in depth clouds concepts and therefore conclude by <strong>achieving all certifications</strong>. With courage and determination I succeeded my plan.</p><p>In the mid 2023, I launched a vast campain to renew all of them in order to be eligible to obtain my aws golden jacket. Now they’re mostly valid up 2026. Not to mention, along with renewal, I prepared certifications from others provider such as google and Azure.</p><figure><img alt="" src="https://cdn-images-1.medium.com/proxy/1*BAUbQAHDhIlYxpMZeFK8Rw.png" /></figure><p><strong>Pyramidal approach and followed path</strong></p><p>The strategy I followed to achieve all of them was going bottom-up, which I think is better for strengthening fundamental building blocks at the associate level before jumping to the pro and specialty level. Not to mention, I deliberately chose to end with the Cloud Practitioner certification. Overall, the path I followed was going through all associate certification levels, then the two professional levels, after which I started all specialty levels and terminated with the Cloud Practitioner certification</p><p>For details, I chose to start with the AWS Solution Architect Associate certification. I diligently followed the course and passed the exam at a Pearson VUE center in Paris. This was the catalyst for me to pursue other certifications. I shared my success with my company, which reimbursed my fees and recognized my achievement.</p><p>Next, I decided to pursue the AWS Developer Associate certification, as I felt confident with my developer background. I continued my research, came across training materials, videos on YouTube, and online training sites such as Whizlabs, Udemy, Tutorials Dojo, Cloud Guru, and Pluralsight. With some preparation, I sat for the exam and succeeded. I found it to be simpler than the Associate certification.</p><p>Following that, I embarked on the AWS SysOps certification. I began my preparation, as I did not have much knowledge in operations and system administration. It was slightly more challenging than the previous two Associate certifications, in my opinion.</p><p><strong>To summarize my journey</strong>, I created a small graph, which I call a “labyrinth,” to illustrate the path I took to achieve my goal. I deliberately used colors, with red emphasizing the difficulty I faced in linking two vertices (certifications).</p><p>One of the main advantages of AWS training and certification is that as long as you are succeeding, you are being offered a 50% voucher on your next exam, which is really encouraging.</p><p><strong>Disclaimer</strong>: I’m not promoting any of the cited companies, but they were useful for me to prepare my courses and also assess my level of preparation before sitting for exams.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*B28kidfUXPJZvIn6uiRzsg.png" /><figcaption>Aws graph labyrinth</figcaption></figure><h4><strong>Preparation plan</strong></h4><p>Bunch of good practices I went through and that might be useful to share for exam takers were:</p><p>1. <strong>Always dive into the official documentation</strong>, read the domains you are being evaluated upon and each task per domain. They are very useful to get familiar with what to expect as questions from the exam.<br>2. <strong>Connect to the AWS Skillbuilders portal</strong>, search the learning plan for your chosen certification. Follow the courses and quizzes.<br>3. At the end of the course, you’ll have <strong>official practice question sets</strong> (usually 20 questions) covering domains and tasks.<br>4. You also have 10 sample questions on the <strong>AWS certification portal to assess your readiness</strong>.<br>5. From there you can start hovering over other practices from <strong>Udemy, Whizlabs, Cloud Guru, Tutorials Dojo, Pluralsight</strong>…</p><h4><strong>Some recipes tackling exam</strong></h4><p>Be mentally prepared for online proctoring during specialty and professional exams. The duration is 170–180 minutes. It’s comforting taking an exam from home, but there are no breaks, and you should not be out of view of your camera.</p><p>Mostly, you might be sitting for the entire duration of the exam, whereas it happened for some exams where you completed an hour in advance.</p><p><strong>Spend time reviewing your answers because in some questions</strong>, you almost have to choose between two good answers.</p><h4>Shortcuts to resources for certifications</h4><p>Following this preparation plan, I’ve gathered from the <strong>AWS official certification page shortcuts to resources</strong> to prepare for different exams. Along with those where you will be understanding the domains and contents of the exam, my advice is to go with other sites and exercise with their practice (Whizlabs, Udemy, Tutorials Dojo, etc.). I don’t cover aws data analytics which wil be retired soon and aws data enginer which is in beta at this time.</p><h4><strong>1- AWS Cloud Practitionner</strong></h4><p>CPP is a foundation to get an overview of what the AWS building block services are, such as S3, EC2, RDS, managed vs. unmanaged services, pricing, billing, models of payment, pay-what-you-use, subscription, spot pricing, and also have knowledge on <strong>AWS pillars of well-architecting framework</strong>.</p><p>Below are the resources you can follow along with to be well prepared for the certification.</p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-cloud-practitioner/AWS-Certified-Cloud-Practitioner_Exam-Guide.pdf">Exam guideline</a></p><p>❇️<a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/134/aws-cloud-practitioner-essentials?trk=5bbfa8ed-a3a8-4256-9bd8-2e02275dde32&amp;sc_channel=el"> Cloud practionner essentials</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/11458/aws-cloud-quest-cloud-practitioner?trk=5bbfa8ed-a3a8-4256-9bd8-2e02275dde32&amp;sc_channel=el">Cloud quest cloud practitionner</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/14050/aws-certified-cloud-practitioner-official-practice-question-set-clf-c02-english?trk=5bbfa8ed-a3a8-4256-9bd8-2e02275dde32&amp;sc_channel=el">Cloud practitionner official practice question</a></p><p>❇️ <a href="http://aws-certified-cloud-practitioner_sample-question/">Cloud practionner samples questions</a></p><h4><strong>2. AWS Solution Associate</strong></h4><p>In other words, compared to CP, SAA will be more comprehensive with a wider range of services. You need to learn more services in order to design your system considering the pillars of well-architected frameworks: <strong>security, performance, resiliency, scalability, high availability, cost, and sustainability</strong>. This will help to strengthen your understanding of cloud concepts and architecture principles.</p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-sa-assoc/AWS-Certified-Solutions-Architect-Associate_Exam-Guide.pdf">Exam guide</a></p><p>❇️<a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/13360/aws-certified-solutions-architect-associate-practice-question-set-saa-c03-french?saa=sec&amp;sec=prep">Practice question-set-saa-c03</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/125/exam-readiness-aws-certified-solutions-architect-associate-digital?saa=sec&amp;sec=prep">Exam-readiness-aws-certified-solutions</a></p><h4><strong>3. AWS Developer Associate</strong></h4><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-dev-associate/AWS-Certified-Developer-Associate_Exam-Guide.pdf">Exam guide</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-dev-associate/AWS-Certified-Developer-Associate_Sample-Questions.pdf">Sample questions</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/13757/aws-certified-developer-associate-official-question-set-dva-c02-english">Oficial practice question set</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/14724/exam-prep-aws-certified-developer-associate-dva-c02">Exam prep course</a></p><h4><strong>4. AWS SysOps Administrator Associate</strong></h4><p>❇️<a href="https://d1.awsstatic.com/training-and-certification/docs-sysops-associate/AWS-Certified-SysOps-Administrator-Associate_Exam-Guide.pdf"> Exam guide</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-sysops-associate/AWS-Certified-SysOps-Administrator-Associate_Sample-Questions.pdf">Sample question</a></p><p>❇️<a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/12485/aws-certified-sysops-administrator-associate-practice-question-set-soa-c02-english?syops=sec&amp;sec=prep"> Official practice question set</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/9313/exam-prep-aws-certified-sysops-administrator-associate">Exam prep course</a></p><h4><strong>5. AWS Certified Solution Architect Professional</strong></h4><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-sa-pro/AWS-Certified-Solutions-Architect-Professional_Exam-Guide.pdf">Exam guide</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-sa-pro/AWS-Certified-Solutions-Architect-Professional_Sample-Questions.pdf">Sample question</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/13270/aws-certified-solutions-architect-professional-official-practice-question-set-sap-c02-english">Official practice question set</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/14951/exam-prep-aws-certified-solutions-architect-professional-sap-c02">Exam prep course</a></p><h4><strong>6. AWS Certified Devops Engineer Professional</strong></h4><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-devops-pro/AWS-Certified-DevOps-Engineer-Professional_Exam-Guide.pdf">Exam guide</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/14673/aws-certified-devops-engineer-professional-official-practice-question-set-dop-c02-english">Official practice question set</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-devops-pro/AWS-Certified-DevOps-Engineer-Professional_Sample-Questions.pdf">Sample question</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/16352/exam-prep-standard-course-aws-certified-devops-engineer-professional-dop-c02-english">Exam prep course</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/16512/exam-prep-enhanced-course-aws-certified-devops-engineer-professional-dop-c02-english">Enhanced prep course</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/18307/aws-jam-journey-security-reinvent-2023">Security Jam Journey</a></p><p>7. <strong>AWS Certified Security Specialist</strong></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/15222/aws-certified-security-specialty-official-practice-question-set-scs-c02-english">Exam guide</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/7636/cloud-quest">Cloud quest security role</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/48/aws-security-fundamentals-second-edition">Aws security fundamental</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/18291/exam-prep-standard-course-aws-certified-security-specialty-scs-c02-english">Examen prep course</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/18292/exam-prep-enhanced-course-aws-certified-security-specialty-scs-c02-english">Enhanced prep course</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/15719/aws-certified-security-specialty-official-practice-exam-scs-c02-english">Official practice question set</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs/AWS_Certified_Security_Specialty_Sample_Questions.pdf">Sample question</a></p><h4><strong>8. AWS Certified Avanced Networking</strong></h4><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-advnetworking-spec/AWS-Certified-Advanced-Networking-Specialty_Exam-Guide.pdf">Exam guide</a></p><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-advnetworking-spec/AWS-Certified-Advanced-Networking-Specialty_Sample-Questions.pdf">Sample question</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/14773/exam-prep-enhanced-course-aws-certified-advanced-networking-specialty-ans-c01">Enhanced exam prep course</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/16957/networking-core-pre-assessment">Network core pre-assesment</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/14434/exam-prep-standard-course-aws-certified-advanced-networking-specialty-ans-c01">Exam prep course</a></p><p>️❇️ <a href="https://explore.skillbuilder.aws/learn/course/16484/aws-cloud-quest-networking">Cloudquest networking role</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/12676/exam-prep-official-practice-question-set-aws-certified-advanced-networking-specialty-ans-c01-english">Official practice question set</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/16879/networking-core-knowledge-badge-assessment">Networking Core Knowledge badge</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/14199/aws-jam-journey-networking">Networking Jam Journey</a></p><h4><strong>9. AWS Certified Database Speciality</strong></h4><p>❇️ ️<a href="https://d1.awsstatic.com/training-and-certification/docs-database-specialty/AWS-Certified-Database-Specialty_Exam-Guide.pdf">Exam guide</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/116/exam-readiness-aws-certified-database-specialty">Exam readiness Database</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/9405/exam-prep-official-practice-question-set-aws-certified-database-specialty-dbs-c01-english">Official practice question set</a></p><p>❇️ ️<a href="https://d1.awsstatic.com/training-and-certification/docs-database-specialty/AWS-Certified-Database-Specialty_Sample-Questions_v1.1.pdf">Sample question</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/14659/aws-jam-journey-database">Database Jam Journey</a></p><h4><strong>10. AWS Certified Machine Learning Speciality</strong></h4><p>❇️ <a href="https://d1.awsstatic.com/training-and-certification/docs-ml/AWS-Certified-Machine-Learning-Specialty_Exam-Guide.pdf">Exam guide</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/12469/exam-prep-official-practice-question-set-aws-certified-machine-learning-specialty-mls-c01-english">Official practice question set</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/27/Exam%2520Readiness%253A%2520AWS%2520Certified%2520Machine%2520Learning%2520-%2520Specialty">Examen readiness course</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/15596/aws-cloud-quest-machine-learning">Cloud qusest Machine learning role</a></p><p>❇️ ️<a href="https://d1.awsstatic.com/training-and-certification/docs-ml/AWS-Certified-Machine-Learning-Specialty_Sample-Questions.pdf">Sample question</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/373/machine-learning-exam-basics">Machine learning basics</a></p><p>❇️ <a href="https://explore.skillbuilder.aws/learn/course/internal/view/elearning/14594/aws-jam-journey-machine-learning">Machine learning Jam Journey</a></p><h4><strong>11. AWS Certified SAP on AWS</strong></h4><p>❇️ ️<a href="https://d1.awsstatic.com/training-and-certification/docs-sap-on-aws-specialty/SAP-on-AWS-Specialty_Exam-Guide.pdf">Exam guide</a></p><p>❇️ ️<a href="https://d1.awsstatic.com/training-and-certification/docs-sap-on-aws-specialty/SAP-on-AWS-Specialty_Sample-Questions.pdf">Sample question</a></p><p>❇️ ️<a href="https://explore.skillbuilder.aws/learn/course/external/view/elearning/12474/aws-certified-sap-on-aws-specialty-official-question-set-pas-c01-english?pas=sec&amp;sec=prep">Official practice question set</a></p><p>❇️ <a href="https://aws.amazon.com/sap/docs/?pas=sec&amp;sec=prep">SAP on AWS</a></p><h4><strong>Some useful links get up to date on aws</strong></h4><p><a href="https://docs.aws.amazon.com/">Aws Documentation</a></p><p><a href="https://aws.amazon.com/whitepapers/">AWS Whitepapers</a></p><p><a href="https://forums.aws.amazon.com/">AWS Forums</a></p><p><a href="https://aws.amazon.com/events/">AWS Events</a></p><p><a href="https://aws.amazon.com/training/">AWS Training and Certification</a></p><p><a href="https://aws.amazon.com/fr/solutions/case-studies/?customer-references-cards.sort-by=item.additionalFields.sortDate&amp;customer-references-cards.sort-order=desc&amp;awsf.language=language%23french&amp;awsf.content-type=*all&amp;awsf.customer-references-location=location%23emea&amp;awsf.customer-references-segment=*all&amp;awsf.customer-references-industry=*all&amp;awsf.customer-references-use-case=*all&amp;awsf.customer-references-category=*all&amp;awsf.customer-references-product=*all">Aws customer success stories</a></p><p><a href="https://reinvent.awsevents.com/?nc2=h_ql_reinv">Aws ReInvent</a></p><p><a href="https://aws.amazon.com/fr/blogs/?awsf.blog-master-category=*all&amp;awsf.blog-master-learning-levels=*all&amp;awsf.blog-master-industry=*all&amp;awsf.blog-master-analytics-products=*all&amp;awsf.blog-master-artificial-intelligence=*all&amp;awsf.blog-master-aws-cloud-financial-management=*all&amp;awsf.blog-master-blockchain=*all&amp;awsf.blog-master-business-applications=*all&amp;awsf.blog-master-compute=*all&amp;awsf.blog-master-customer-enablement=*all&amp;awsf.blog-master-customer-engagement=*all&amp;awsf.blog-master-database=*all&amp;awsf.blog-master-developer-tools=*all&amp;awsf.blog-master-devops=*all&amp;awsf.blog-master-end-user-computing=*all&amp;awsf.blog-master-mobile=*all&amp;awsf.blog-master-iot=*all&amp;awsf.blog-master-management-governance=*all&amp;awsf.blog-master-media-services=*all&amp;awsf.blog-master-migration-transfer=*all&amp;awsf.blog-master-migration-solutions=*all&amp;awsf.blog-master-networking-content-delivery=*all&amp;awsf.blog-master-programming-language=*all&amp;awsf.blog-master-sector=*all&amp;awsf.blog-master-security=*all&amp;awsf.blog-master-storage=*all">Blog Aws</a></p><p><a href="https://aws.amazon.com/fr/faqs/">Aws FAQ</a></p><h4><strong>My exam records</strong></h4><p>I’m happy to have succeeded in all my exams. I’m still continuing learning. <strong>Learning is sharing, self-motivating, linking with community, being rigorous, and having time-set goals.</strong> I’m thankful to <strong>those who inspire me in this journey, and I’ll be happy if I could do the same for future learners</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/1*H9zhdrnKYwH19gBUAuv3XA.jpeg" /><figcaption>My aws certification record</figcaption></figure><h4><strong>AWS Golden Jacket</strong></h4><p>When I completed the first round of <strong>AWS 11 certifications</strong>, I thought I would be granted a golden jacket automatically if I claimed it. Unfortunately, it was not the case and it unraveled the long process. Your company has to be an AWS partner to claim it on your behalf. Anyway, I kept learning AWS cloud out of passion, including <strong>free badges, AWS educate badges, and AWS .NET badges</strong>.</p><p>In November-December 2023, I decided to attend AWS re:Invent. It became a big opportunity attending this tech event. I learned and met people and resources in the field.</p><h4><strong>Conclusion</strong></h4><p>After completing all these certifications in depth, I understood that <strong>the cloud has played and will play an important role in the evolution of business services</strong>, specifically in areas such as the main pillars: scalability, availability, managed services, automation, security, performance, cost, and sustainability.</p><p>Although a majority of businesses have already started their migration to the cloud, technologies such as generative intelligence and quantum computing will appear in the near future to <strong>bring dazzling and revolutionary developments to our daily lives</strong>.</p><p>Services like AWS Bedrock, Amazon Bracket, and Amazon Quantum Exposed at recent AWS Reinvent illustrate this trend.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ea0d0c653d65" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>