<?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 Kamlesh Mali on Medium]]></title>
        <description><![CDATA[Stories by Kamlesh Mali on Medium]]></description>
        <link>https://medium.com/@kamleshmali81?source=rss-2b472a463b28------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/0*OyJjbKvaa9lNwTy7</url>
            <title>Stories by Kamlesh Mali on Medium</title>
            <link>https://medium.com/@kamleshmali81?source=rss-2b472a463b28------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 23:24:38 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@kamleshmali81/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[We stopped asking AI questions. Now we’re handing it the keys…]]></title>
            <link>https://medium.com/@kamleshmali81/we-stopped-asking-ai-questions-now-were-handing-it-the-keys-b76c9045ec93?source=rss-2b472a463b28------2</link>
            <guid isPermaLink="false">https://medium.com/p/b76c9045ec93</guid>
            <category><![CDATA[enterprise-ai-solutions]]></category>
            <category><![CDATA[ai-governance]]></category>
            <category><![CDATA[agentic-ai-systems]]></category>
            <category><![CDATA[engineering-leadership]]></category>
            <dc:creator><![CDATA[Kamlesh Mali]]></dc:creator>
            <pubDate>Tue, 27 Oct 2026 05:08:50 GMT</pubDate>
            <atom:updated>2026-10-27T05:16:10.137Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>In the last seven weeks, AI quietly crossed a line.</strong></p><p>For three years we asked chatbots questions and then did the work ourselves. Now the leading labs have shipped agents that do the work for us.</p><p>These always-on agents run on their own cloud computers, keep going after we close the app, and act across our email, calendar, browser and business systems.</p><p>The shift is happening along several lines at once.</p><p>Agents are moving from reactive to proactive:</p><p>· They brief us in the morning,</p><p>· Flag what needs a decision, and</p><p>· Handle the rest in the background.</p><p><strong>They are moving from borrowing our logins to holding identities of their own, with separate credentials, audit trails and spending limits.</strong></p><p><strong>And the competition has moved from capability to trust.</strong></p><p>Models are now close in raw ability, so what separates products is</p><p>· How they handle credentials,</p><p>· When they stop to ask permission, and</p><p>· How they fail.</p><p>The field already has distinct personalities from leading labs</p><p>Ø OpenClaw started the category as an open-source, self-hosted agent that works with any model, and an independent foundation now steers it toward enterprise use.</p><p>Ø Meta’s Muse brings agents to consumers at scale through WhatsApp, giving each user a dedicated secure virtual machine, though its early privacy incidents show what fast shipping costs.</p><p>Ø Instinct, a small startup valued at ten billion dollars within months, lets you simply text or call your agent, and its terms state plainly that the agent acts as you.</p><p>Ø Anthropic’s Claude Duet is aimed at professionals, with daily briefings, approval controls and an architecture where the agent never sees your passwords.</p><p>Ø OpenAI’s dots live inside ChatGPT and reach thousands of apps.</p><p>Ø Google gives every Gemini work agent its own role-scoped identity and activity log.</p><p><strong>For an engineering enterprise, the right response is neither to chase every launch nor to wait.</strong></p><p>· Choose an architecture before a brand: self-hosted for regulated and isolated environments, hosted where speed matters, and agent-owned identities wherever auditability counts.</p><p>· Treat agents as non-human identities with least-privilege access. Start in read-only and draft-first modes, and grant autonomy only as trust is earned.</p><p>· Build a curated catalog of approved skills and connectors.</p><p>· Test for prompt injection before anything reaches production. And</p><p>· Make your own systems ready for the moment your customers’ agents start calling, emailing and transacting with you.</p><p>Done well, the payoff is substantial.</p><p>· Engineers get back the hours lost to coordination and status chasing.</p><p>· Delivery operations run around the clock.</p><p>· Decisions start with the context already gathered.</p><p>· Agent skills compound across teams.</p><p>· And sanctioned, observable agents replace the shadow AI that is already spreading through most organizations.</p><p><strong>Software has always waited for a human to click. That assumption is ending.</strong></p><p>Agents don’t wait to be asked. They notice, prepare and act, and they come back to us only when a decision truly needs a human.</p><p>The systems we build will have to change with it. They must be designed for two kinds of users,</p><p>· People who need clarity and control, and</p><p>· Machines that need clean interfaces, clear permissions and a full record of what they did and why.</p><p>The enterprises that win will not be the ones with the most agents. They will be the ones that can trust their agents to act and can prove it.</p><p><strong>The enterprises that win will not be the ones with the most agents. They will be the ones that can trust their agents to act, and prove it.</strong></p><p>#AgenticAI #EnterpriseAI #EngineeringLeadership #AIGovernance</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=b76c9045ec93" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Foundations of Agentic AI]]></title>
            <link>https://medium.com/@kamleshmali81/the-foundations-of-agentic-ai-acbfe732bf20?source=rss-2b472a463b28------2</link>
            <guid isPermaLink="false">https://medium.com/p/acbfe732bf20</guid>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[generative-ai-tools]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[agentic-ai]]></category>
            <dc:creator><![CDATA[Kamlesh Mali]]></dc:creator>
            <pubDate>Sun, 17 Aug 2025 03:56:12 GMT</pubDate>
            <atom:updated>2025-08-21T08:53:20.879Z</atom:updated>
            <content:encoded><![CDATA[<h3>Foundation of Agentic AI</h3><p>Artificial Intelligence has always aimed to <strong>extend human intelligence through machines</strong>. Early AI was largely passive, you gave it inputs, it returned outputs. Today, with <strong>Agentic AI</strong>, we move into an era where systems can <strong>perceive goals, reason about actions, and autonomously execute tasks</strong> without constant human nudging.</p><p>This is a <strong>paradigm shift</strong> in enterprise AI adoption:</p><ul><li><strong>Predictive AI (2000s):</strong> ML models predicting outcomes, fraud detection, recommendation systems, churn prediction.</li><li><strong>Generative AI (2020+):</strong> LLMs and diffusion models generating text, images, or code on demand.</li><li><strong>Agentic AI (Now):</strong> Autonomously pursuing objectives, coordinating workflows, and using tools, effectively becoming <strong>digital colleagues</strong>.</li></ul><p>The “agent” abstraction is powerful because it combines <strong>intelligence + autonomy + action</strong>. Rather than being just a conversational interface, an agent can <strong>drive real-world outcomes</strong> across finance, IT, customer support, or R&amp;D.</p><p>👉 This represents not just a technology adoption, but a <strong>new systems architecture layer</strong>.</p><ol><li><strong>Defining an Agent: Beyond Chatbots</strong></li></ol><p>An <strong>agent</strong> can be thought of as a <strong>closed-loop software entity</strong> that:</p><ol><li><strong>Perceives</strong> : takes in goals, user prompts, system events, or environmental signals.</li><li><strong>Reasons</strong> : decides what needs to happen next, often decomposing complex goals into smaller steps.</li><li><strong>Acts</strong> : executes actions via APIs, databases, tools, or even interacting with other agents.</li><li><strong>Learns : </strong>updates memory, refines strategies, and adapts from outcomes.</li></ol><p>This is known as the <strong>Perception–Reasoning–Action (PRA) loop</strong>, often extended with memory and guardrails.</p><p>Contrast this with a <strong>traditional chatbot</strong>:</p><ul><li><strong>Chatbot:</strong> You ask a question → it answers (no memory, no autonomy, no execution).</li><li><strong>Agent:</strong> You assign a goal → it figures out the path, calls the right tools, monitors progress, and adapts.</li></ul><p>📌 Canonical frameworks shaping this space:</p><ul><li><strong>ReAct (Reason + Act):</strong> Interleaves “thinking” and “tool calls” to ground reasoning in reality.</li><li><strong>Reflexion:</strong> Agents reflect on mistakes, critique themselves, and improve.</li><li><strong>ReWOO (Plan–Work–Solve):</strong> Separates planning from execution for better efficiency.</li></ul><p><strong>2. Historical Journey: How We Arrived Here</strong></p><p>To understand Agentic AI, it helps to trace its lineage:</p><ul><li><strong>1950s–1980s (Symbolic AI):</strong><br> Early agents like <strong>SHRDLU</strong> (natural language in a blocks world) and <strong>ELIZA</strong> (rule-based dialogue). Intelligent but brittle, with no adaptability.</li><li><strong>1990s–2000s (Software Agents):</strong><br> Multi-agent systems in distributed computing. JADE and other frameworks allowed basic collaboration and task automation.</li><li><strong>2010s (Deep Learning &amp; Reinforcement Learning):</strong><br> Breakthroughs like <strong>AlphaGo</strong> and <strong>OpenAI Five</strong> showed agents could learn strategies, plan moves, and adapt through reinforcement learning.</li><li><strong>2022–Now (LLM-driven Agents):</strong><br> With GPT-4, Claude, and Gemini, agents gained <strong>general-purpose reasoning and language skills</strong>. Frameworks like <strong>LangChain, AutoGen, CrewAI, and Swarm</strong> made it possible to wrap LLMs in full agent loops.</li></ul><p>👉 What’s new isn’t the <em>idea</em> of agents, but their <strong>practical usability at scale</strong>. With LLMs, agents are no longer research toys, they are <strong>enterprise-ready actors</strong>.</p><p><strong>3. Core Building Blocks of Agentic AI</strong></p><p>Every modern agentic system, whether powering a personal assistant or an enterprise automation network, is built on the same building blocks.</p><p><strong>3.1 Planner / Policy</strong></p><ul><li><strong>The decision-making brain</strong>.</li><li>Variants:</li></ul><p><strong>ReAct</strong> → Reason + Act loop.</p><p><strong>Reflexion</strong> → Adds self-feedback.</p><p><strong>ReWOO</strong> → Plan first, then fetch and solve.</p><p><strong>3.2 Tooling Layer</strong></p><ul><li>Agents can’t manipulate the world directly, they <strong>call tools</strong>.</li><li>Tools are exposed as structured function calls with JSON I/O contracts.</li><li>Examples: Search engines, CRMs, ERP APIs, ticketing systems, code execution.</li></ul><p><strong>3.3 Memory</strong></p><ul><li><strong>Short-term memory:</strong> A scratchpad for reasoning steps.</li><li><strong>Long-term memory:</strong> Vector databases (pgvector, Pinecone, Weaviate) with RAG (retrieval-augmented generation).</li><li><strong>Skill libraries:</strong> Persistent knowledge of reusable strategies (e.g., Voyager’s skill accumulation).</li></ul><p><strong>3.4 Orchestrator</strong></p><ul><li>The “conductor” of the agent system.</li><li>Options: <strong>Single agent loop</strong> or <strong>multi-agent graph orchestration</strong> (LangGraph, AutoGen, CrewAI, Swarm, smolagents).</li></ul><p><strong>3.5 Observation / Environment</strong></p><ul><li>Agents need to perceive environments: files, web, terminals, APIs, or applications.</li><li>Benchmarks like <strong>WebArena, AgentBench, SWE-bench</strong> test competence.</li></ul><p><strong>3.6 Guardrails &amp; Safety</strong></p><ul><li>Policies prevent unintended consequences.</li><li>Allow/deny lists, pre-execution validation, undo/damage confinement.</li><li>Tools: <strong>NeMo Guardrails, Gorilla GoEX, RAIL-style policies</strong>.</li></ul><p><strong>3.7 Observability</strong></p><ul><li>Logging, traces, dashboards, and evals (LangSmith-style).</li><li>Essential for debugging, measuring, and governing agents.</li></ul><p><strong>4. Canonical Architectures</strong></p><p>Three <strong>reference architectures</strong> dominate today:</p><p><strong>(A) ReAct, Single Agent with Tool Use</strong></p><ul><li>Interleaves reasoning and acting.</li><li>Example: An agent solving a research task by alternating between “thinking” and “Google search calls.”</li></ul><p><strong>(B) Supervisor–Worker Multi-Agent Pattern</strong></p><ul><li>Supervisor agent applies business policy.</li><li>Routes tasks to domain agents (Developer, Data, CRM).</li><li>Shared memory enables collaboration.</li><li>Used in <strong>LangGraph, AutoGen, CrewAI</strong> ecosystems.</li></ul><p><strong>C) ReWOO (Plan–Work–Solve)</strong></p><ul><li>Drafts a <strong>plan</strong> before execution.</li><li>Executes tool calls in parallel.</li><li>Integrates results in a final solve step.</li><li>Benefits: Token efficiency, robustness, reduced costs.</li></ul><p><strong>5. The Architecture Lens</strong></p><p>The biggest question is <strong>where Agentic AI fits in enterprise stacks</strong>.</p><p>Think of it as a <strong>cognitive layer above your systems</strong>:</p><ul><li><strong>Systems of Record (ERP, CRM, ITSM):</strong> Contain structured enterprise data.</li><li><strong>Systems of Engagement (chat, dashboards, portals):</strong> User-facing interfaces.</li><li><strong>Agentic AI Layer (new):</strong> Sits above both, orchestrating across them with reasoning and tool use.</li></ul><p><strong>Example: IT Ops Agent</strong></p><ol><li>System event: “Database latency rising.”</li><li>Agent perceives → queries monitoring tools.</li><li>Reasons → suspects overloaded queries.</li><li>Acts → generates a mitigation playbook or scales cluster automatically.</li></ol><p>This <strong>event-driven autonomy</strong> reduces manual toil and speeds incident response.</p><p>📌 Enterprise Implications:</p><ul><li><strong>Separation of concerns:</strong> Orchestrator handles goals, agents handle tasks, tools handle integrations.</li><li><strong>Scalability path:</strong> Start small (e.g., RFP summarizer), scale to ecosystems.</li><li><strong>Governance:</strong> Who approves actions? How to log reasoning? How to sandbox?</li></ul><p><strong>6. Cheat-Sheet: Orchestration Stacks</strong></p><p>Choosing an orchestration framework depends on your enterprise needs:</p><ul><li><strong>LangGraph:</strong> Robust stateful orchestration, retries, human-in-loop.</li><li><strong>OpenAI Swarm:</strong> Lightweight, ergonomic, Python-native.</li><li><strong>AutoGen/AG2:</strong> Research-friendly, mature conversation patterns.</li><li><strong>CrewAI:</strong> Role-based teams for fast prototyping.</li><li><strong>smolagents:</strong> Minimalist, easy to audit, good for compliance-heavy orgs.</li></ul><p><strong>7. Evaluation &amp; Metrics</strong></p><p>Enterprises must evaluate agents with the rigor of product testing:</p><ul><li><strong>Benchmarks:</strong> WebArena (web tasks), AgentBench (multi-domain), SWE-bench (coding).</li><li><strong>Metrics:</strong> Success rate, cost per episode, intervention rate, latency, safety violations.</li><li><strong>Continuous evaluation:</strong> Log every tool call, curate a “red set” of failures, replay regularly.</li></ul><p><strong>8. Safety &amp; Risk Containment</strong></p><p>Trust is non-negotiable for enterprise adoption.</p><ul><li><strong>Before acting:</strong> Dry-run simulation, allow-lists, approval gates.</li><li><strong>During runtime:</strong> Guardrails, policy enforcement, damage confinement.</li><li><strong>After action:</strong> Immutable logs, incident playbooks, compliance auditability.</li></ul><p><strong>9. References &amp; Suggested Reading</strong></p><ul><li>🎥 <strong>Andrej Karpathy</strong> : <em>The Rise of Agentic AI</em></li><li>📄 <strong>Russell &amp; Norvig</strong> : <em>Artificial Intelligence: A Modern Approach</em> (Chapter 2: Intelligent Agents)</li><li>📄 <strong>ReAct Paper</strong> : <em>Synergizing Reasoning and Acting in Language Models</em></li><li>📄 <strong>Lewis et al., 2020: </strong>Retrieval-Augmented Generation (RAG)</li><li>Framework Docs: <strong>LangChain, AutoGen, CrewAI, Swarm, smolagents</strong></li><li>Benchmarks: <strong>WebArena, AgentBench, SWE-bench</strong></li><li>Blog: <strong>Lilian Weng (OpenAI)</strong> : A Brief History of Autonomous Agents</li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=acbfe732bf20" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ Where Shiva Waited ]]></title>
            <link>https://medium.com/@kamleshmali81/where-shiva-waited-5a013dee3503?source=rss-2b472a463b28------2</link>
            <guid isPermaLink="false">https://medium.com/p/5a013dee3503</guid>
            <category><![CDATA[bholenath]]></category>
            <category><![CDATA[india]]></category>
            <category><![CDATA[sanatana-dharma]]></category>
            <category><![CDATA[lord-shiva]]></category>
            <dc:creator><![CDATA[Kamlesh Mali]]></dc:creator>
            <pubDate>Fri, 25 Jul 2025 16:31:27 GMT</pubDate>
            <atom:updated>2025-07-25T17:08:04.931Z</atom:updated>
            <content:encoded><![CDATA[<p>(My sacred encounter, remembered in devotion)</p><p>I was not seeking,</p><p>not praying,</p><p>not even thinking of Him.</p><p>And still,</p><p>He found me.</p><p>Not in thunder,</p><p>nor in flame,</p><p>but in the soft steps of a 7 year child,</p><p>wearing green,</p><p>like spring wrapped in silence.</p><p>At Bhagwati temple on the other day,</p><p>he came near,</p><p>eyes like still water,</p><p>and said,</p><p>“Have you seen them all?”</p><p>“Seen who?” I asked, unsure.</p><p>He pointed to the stones,</p><p>twelve sacred forms,</p><p>lit by sunlight through peepal leaves.</p><p>“These are the Jyotirlingas,” he said,</p><p>“pillars of light where Shiva breathes.</p><p>You have walked among seven,</p><p>but five still wait.</p><p>And they do not wait in stone alone,</p><p>they wait in you.”</p><p>My breath held like a frozen bell.</p><p>I turned to ask his name.</p><p>He smiled.</p><p>“I am a Poojari of Shiva.</p><p>Nothing more, nothing less.”</p><p>Then he touched a Shivling near the edge,</p><p>forgotten, cracked,</p><p>and said,</p><p>“This one is dearest to me.”</p><p>No grandeur, no gold, just stillness.</p><p>Then,</p><p>he walked away</p><p>like mist from a morning lamp.</p><p>That night,</p><p>the temple was quieter than usual.</p><p>No birds,</p><p>no wind.</p><p>Just silence deeper than thought.</p><p>And by morning,</p><p>a green serpent lay coiled</p><p>around the very Shivling</p><p>he had touched.</p><p>Still.</p><p>Watching.</p><p>Alive.</p><p>Time itself seemed to kneel.</p><p>Villagers whispered,</p><p>devotees gathered.</p><p>But I stood at a distance,</p><p>not as a witness,</p><p>but as one awakened.</p><p>Since that day,</p><p>I no longer seek Shiva</p><p>only in the Himalayas or temples.</p><p>I look for Him</p><p>in every forgotten corner,</p><p>every innocent voice,</p><p>every cracked stone</p><p>and serpent’s gaze.</p><p>Because now I know,</p><p>the Jyotirlingas are not places.</p><p>They are thresholds,</p><p>twelve inner flames</p><p>lit by grace,</p><p>one by one.</p><p>And now I walk,</p><p>not just to temples,</p><p>but inward,</p><p>toward the One</p><p>who came to me</p><p>as a boy in green,</p><p>and asked a question</p><p>that awakened my soul:</p><p>“When will you complete the journey?”</p><p>– – – – – –</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5a013dee3503" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Krishna Walked Beside Me in Dwarka]]></title>
            <link>https://medium.com/@kamleshmali81/krishna-walked-beside-me-in-dwarka-74162d7d0cfe?source=rss-2b472a463b28------2</link>
            <guid isPermaLink="false">https://medium.com/p/74162d7d0cfe</guid>
            <category><![CDATA[devotional]]></category>
            <category><![CDATA[sanatana-dharma]]></category>
            <category><![CDATA[bhagwat-gita]]></category>
            <category><![CDATA[indian]]></category>
            <category><![CDATA[india]]></category>
            <dc:creator><![CDATA[Kamlesh Mali]]></dc:creator>
            <pubDate>Fri, 25 Jul 2025 16:08:24 GMT</pubDate>
            <atom:updated>2025-07-25T16:09:43.296Z</atom:updated>
            <content:encoded><![CDATA[<p>(My Journey with Krishna at Dwarka – real incident)</p><p>He came to find me, not in dreams,</p><p>But on the platform midst worldly scenes.</p><p>At Veraval, where my journey began,</p><p>He stood a simple, smiling man.</p><p>No peacock crown or flute in hand,</p><p>But still, I knew this was not bland.</p><p>He called my name, his voice so clear,</p><p>My name was his with his Aadhaar card near.</p><p>We boarded the train, night softly fell,</p><p>Through winding tracks, silence in train would tell,</p><p>Of hidden truths the soul once knew,</p><p>Unspoken things that still felt true.</p><p>He stayed with me through every mile,</p><p>My heart unraveling at every place.</p><p>No trace of pride, no need to show,</p><p>Just care, just grace, in every flow.</p><p>In Dwarka’s streets, he led the way,</p><p>Through temple lights and ocean spray.</p><p>To Bet Dwarka we crossed the tide,</p><p>With waves below and God beside.</p><p>And there, beneath the temple sky,</p><p>He gave me rice, I don’t know why,</p><p>My eyes welled up, my words were few,</p><p>For kings don’t serve, unless love is true.</p><p>The world may doubt, the world may ask,</p><p>But truth wears no worldly mask.</p><p>He walked with me, it was no lore,</p><p>Krishna was mine, and I was his evermore.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=74162d7d0cfe" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Strategizing Digital Transformation 2.0]]></title>
            <link>https://medium.com/@kamleshmali81/strategizing-digital-transformation-2-0-c7762bd81073?source=rss-2b472a463b28------2</link>
            <guid isPermaLink="false">https://medium.com/p/c7762bd81073</guid>
            <category><![CDATA[cio]]></category>
            <category><![CDATA[information-technology]]></category>
            <category><![CDATA[strategy]]></category>
            <category><![CDATA[digital-transformation]]></category>
            <category><![CDATA[strategic-planning]]></category>
            <dc:creator><![CDATA[Kamlesh Mali]]></dc:creator>
            <pubDate>Thu, 10 Feb 2022 10:49:36 GMT</pubDate>
            <atom:updated>2022-02-15T03:49:33.905Z</atom:updated>
            <content:encoded><![CDATA[<p>In this digital innovation era, trends around digitization, digitalization and digital transformation have already made organizations across all business sectors to reckon whether their current way of operation will make them extinct, just survive or thrive in this continuously scaling competitive technical world. Upcoming next wave on digital transformation is definite to swap away the lethargic detractors.</p><h4>Roots in Innovation along with transformation</h4><p>To be in race, well established enterprises need to cultivate and maintain intrapreneurship mindset. Frugality based innovations are already disrupting many market segments which were earlier thought to be least conquerable. Merely claiming to be a cloud native organization, rationalizing application portfolio, modernizing legacy applications, and automating workflows does not qualify as an innovative driven working, the need of hour is to rediscover themselves, be innovative across all business lines, modernize experiences and stay integrated with full agility. Roof-shot and Moon-shot innovations needs to be progressively interwoven across.</p><p>Strategic organizational goal of being a composable enterprise is already mainstreaming now, with focus on resiliency in this disruptive era. Resiliency mandates equipping each business line with technical decision making, to unleash potential of advance technologies in an iterative continuum way and stay ahead instead of just being relevant. Roof-shot innovations are the state of the art here, requiring a business line to keep transforming, rejuvenating and stay aligned with enterprise’s transformation initiative plus constitute as a building block of well composed enterprise.</p><h4>Omni-transformation for business capability multiplier</h4><p>Unique business capabilities across business lines sums up to an enterprise’s overall capabilities. With an enterprise following more complex processes, lesser the scope of cultivating culture of innovation. Periodic retrospections on replacement of mechanical IT with innovative IT catalyzes the probability of developing new differentiated business capabilities or sharpens existing business capabilities. Business applications needs to novel around ideas to build capabilities leveraging Transformational IT instead of Transactional IT. Technology again as an integrator for these newly discovered business capabilities provides fertile grounds for business capabilities multiplication and positions an enterprise in offering more unique business capabilities compared to its competitors.</p><h4>Future digital enterprises</h4><p>Super creative, hyper-connected, fully agile, and faster than yesterday along with seasoned intrapreneurship mindset are the bare minimum requirements to venture on digital transformation 2.0 journey. Since today’s problems can’t be solved using yesterday’s technology, then for becoming a future digital enterprise, how can the old school thoughts prevail?</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c7762bd81073" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>