<?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[Deepflow AI - Medium]]></title>
        <description><![CDATA[Collaborative platform for production-ready data-science - Medium]]></description>
        <link>https://medium.com/deepflow-ai?source=rss----eeca9931d9b---4</link>
        <image>
            <url>https://cdn-images-1.medium.com/proxy/1*TGH72Nnw24QL3iV9IOm4VA.png</url>
            <title>Deepflow AI - Medium</title>
            <link>https://medium.com/deepflow-ai?source=rss----eeca9931d9b---4</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Thu, 19 Nov 2026 01:06:46 GMT</lastBuildDate>
        <atom:link href="https://medium.com/feed/deepflow-ai" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Clean Up Your Data Science Mess]]></title>
            <link>https://medium.com/deepflow-ai/clean-up-your-data-science-mess-781d4c1bd579?source=rss----eeca9931d9b---4</link>
            <guid isPermaLink="false">https://medium.com/p/781d4c1bd579</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[enterprise]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Fabien Durand]]></dc:creator>
            <pubDate>Fri, 10 Jul 2020 13:45:10 GMT</pubDate>
            <atom:updated>2020-07-10T13:45:10.418Z</atom:updated>
            <content:encoded><![CDATA[<p><em>In Data Science, moving from experimentation to production is a challenge. Here’s how you can clean up the mess by implementing DataOps practices. This post was originally published on the </em><a href="https://www.deepflow.ai/clean-up-your-data-science-mess/"><em>Deepflow blog</em></a><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*Jy4-PBae2Gpsc2k9BOTvAw.jpeg" /></figure><p>From experimentation to production the road ahead is not paved. While trial and error is the essence of the Data Scientist’s life, critical AI systems need clarity. For example, if you built a credit scoring AI that automatically grants loans to your customers, you need to understand how it makes decisions about who should or shouldn’t receive a loan. This is the data scientist’s job to ensure transparency, but it’s not self-evident.</p><p>For a data scientist, a typical workflow looks like the this: You find a dataset, explore it in a notebook, try modeling techniques, fail, repeat … until it works. After hours of adding new elements to a project, it’s not uncommon finding oneself with a sprawling monster of data sources, notebooks and CSVs. That’s when you start asking yourself, how do I put that in production? The “mess”, that is an inherent part of the data exploration and modeling process, now prevents for a smooth transition of the project to production. Progress starts to stall and time to production keeps getting longer. To add to the pressure, your boss wants your solution live before the end of the week. But what took months of iterations and back and forth won’t magically turn into production code. It’s a problem that engineers know too well: avoiding technical debt by refactoring legacy code. Messy and hard-to-understand systems are not only a technical challenge, they routinely impact the bottom line of companies big and small (<a href="https://www.deepflow.ai/designing-trustworthy-ai/">read our article on Explainable AI</a>). Needless to say, you need to clean up the mess.</p><figure><img alt="From Jupyter notebook to production, an allegory." src="https://cdn-images-1.medium.com/max/800/0*oGrbtMxsvg17SnEG.jpg" /><figcaption>From Jupyter notebook to production, an allegory.</figcaption></figure><p>The hand-off of projects from “research” scientists to the “production” engineers is more standardized in large companies where R&amp;D is its own entity and processes are in place to facilitate communication. Data teams in small organizations bear the brunt of limited resources and short development cycles which adds even more pressure to avoid “messy” projects.</p><blockquote><em>(In Enterprise AI) The most exciting problems yet to be solved are in the deployment and serving space. One reason for the lack of serving solutions is the lack of communication between researchers and production engineers. At companies that can afford to pursue AI research (e.g. big companies), the research team is separated from the deployment team, and the two teams only communicate via the p-managers: product managers, program managers, project managers. Small companies, whose employees can see the entire stack, are constrained by their immediate product needs. Only a few startups, usually those founded by accomplished researchers with enough funding to hire accomplished engineers, have managed to bridge the gap. These startups are poised to take a big chunk of the AI tooling market. </em><strong><em>Chip Huyen — “</em></strong><a href="https://huyenchip.com/2020/06/22/mlops.html"><strong><em>What I learned from looking at 200 machine learning tools</em></strong></a><strong><em>”</em></strong></blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/480/0*tKQN0AZYUEBPOZQI.gif" /><figcaption>Putting your AI models in production</figcaption></figure><h3>Moving AI in production</h3><p>In the software industry, the practice of managing the communication and hand offs between Developer and IT Operations is called DevOps. In practical terms, DevOps refers to the collection of industry practices that standardize how code is created, delivered, deployed and maintained in a consistent way from start to finish within a same team. It introduces two foundational concepts: Continuous Integration (CI) and Continuous Delivery (CD). While CI automates the building, packaging and testing of applications, CD automates the delivery of the applications to the different environments (e.g. development, testing, production).</p><p>Data Science works in a very similar way. After all, it’s another flavor of software engineering. But because of some of the differences we discussed in the previous part DevOps principles cannot be transposed exactly in the same way. That’s why DataOps is gaining interest in the industry, it goes beyond DevOps to offer a tailored approach to manage Data Science projects and avoid the “mess” from start to finish.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*rU8bo9ov-noW3w5s.png" /><figcaption>DevOps vs. DataOps — <a href="https://datakitchen.io/dataops-cookbook-second-edition.html">The DataOps Cookbook 2nd Ed.</a></figcaption></figure><p>The main reason for DataOps as a separate approach is that contrary to traditional engineering teams, Data Science teams are often composed of many non-engineer types, like analysts or research-focused data scientists. Their skill-set is different from engineers. Their goal is not push production-ready code but to find how data can be leveraged to answer specific problems, explore datasets to find answers, and build predictive models. Their focus is only on improving the accuracy of their models and producing actionable visualizations.</p><p>Another differentiating aspect is the range of tools used. The typical mindset of an engineering team is to create a standardized “sandbox” (isolated development environment to qualify code before pushing it to production) for every engineer in the team. When working on the same projects, every engineer sandbox will look exactly the same to make sure individual changes can be merged easily, thus speeding up development time and improving the quality of the overall code-base. In the context of data teams, creating a standardized environment is much more challenging. Data scientists and analysts tend to use their own tools. The exploratory nature of their work makes it almost a necessity. In this context, we prefer to talk about “orchestration” rather than “building” (see previous graph).</p><p>In conclusion, while DevOps deals with managing the complexity of the engineering life-cycle through centralization (aka. one DevOps team per company), DataOps aims to manage the freedom of Data Science through orchestration (aka. data teams are local and are distributed across the oragnization).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*iqXVt4A2Y7EbrLqj.png" /><figcaption>Continuous Delivery for Machine Learning end-to-end process — <a href="https://martinfowler.com/articles/cd4ml.html#cd4ml-end-to-end.png">CD4ML</a></figcaption></figure><h3>The future of Enterprise AI: business leaders with AI skills</h3><p>As highlighted in the graph below, the biggest change in AI is happening behind the curtain. AI researchers, while still <a href="https://www.nytimes.com/2017/10/22/technology/artificial-intelligence-experts-salaries.html">expensive and in scarce numbers</a>, are not the coveted unicorns they once were.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*5DICjQgHu6dBsSdP.png" /><figcaption><a href="https://www2.deloitte.com/us/en/insights/focus/cognitive-technologies/ai-adoption-in-the-workforce.html">Deloitte Insights — AI in the Enterprise</a></figcaption></figure><p>After years of educating themselves, an increasing number of companies are realizing that AI research is overkill for their business. The already available algorithms and techniques are more than sufficient to tackle existing problems, no need to reinvent the wheel. What is missing is the ability for organizations to understand how to leverage the state of the art into working solutions. Business leaders and managers need to be trained to understand and utilize existing AI solutions.</p><blockquote><em>Many companies rush into the AI race without clear objectives, hope a brilliant AI researcher and a technology team can create something great without guidance, and end up with little to show for it. Recruiting an AI quarterback to provide the business input, and ensuring success with well-defined metrics, is the most important job that most companies miss. </em><strong><em>Beck, Davenport, and Libert — “</em></strong><a href="https://hbr.org/2019/03/the-ai-roles-some-companies-forget-to-fill"><strong><em>The AI roles some companies forget to fill</em></strong></a><strong><em>”.</em></strong></blockquote><h3>Track your data team progress and reduce time to production.</h3><p>With workflows that are production-ready by design, Deepflow makes it easy to manage the versioning, CI/CD, testing and monitoring of your models from conception to execution.</p><h3>&gt;&gt;<a href="https://bit.ly/Deepflow-Early-Adopter">Join the next batch of early adopters</a>&lt;&lt;</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*yVI-uVlUsSD8PXyA.jpg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=781d4c1bd579" width="1" height="1" alt=""><hr><p><a href="https://medium.com/deepflow-ai/clean-up-your-data-science-mess-781d4c1bd579">Clean Up Your Data Science Mess</a> was originally published in <a href="https://medium.com/deepflow-ai">Deepflow AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Designing trustworthy AI applications]]></title>
            <link>https://medium.com/deepflow-ai/designing-trustworthy-ai-applications-1285cb3e9fe8?source=rss----eeca9931d9b---4</link>
            <guid isPermaLink="false">https://medium.com/p/1285cb3e9fe8</guid>
            <category><![CDATA[trust]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[team]]></category>
            <category><![CDATA[enterprise]]></category>
            <dc:creator><![CDATA[Fabien Durand]]></dc:creator>
            <pubDate>Mon, 22 Jun 2020 10:10:35 GMT</pubDate>
            <atom:updated>2020-06-22T10:10:35.237Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Today, designing AIs you and your customers can trust is more than a competitive advantage, it’s a necessity. Let’s look at why and how companies can deliver on that promise. This post was originally published on </em><a href="https://www.deepflow.ai/designing-trustworthy-ai/"><em>the Deepflow blog</em></a><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*U2ynei29WpsANMeCrJH_Tw.jpeg" /></figure><p>How would you feel if one day you woke up only to find yourself declared a criminal by an algorithm?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/325/0*CVfPtlGjNRmuLC_Q.gif" /></figure><p>Well, that’s what <a href="https://undark.org/2020/06/01/michigan-unemployment-fraud-algorithm/">happened to a bunch a people in Michigan</a>. In 2014, an automated system designed to detect and report unemployment fraud started to charge people with misrepresentation and demand repayment. With no human in the driver seat 40,000 people were billed about five times their original benefits. This whole ordeal lasted for two years. The agency in charge later admitted than 93 percent of the charges had been erroneous, but only after completing a partial audit of the system. To this day, most of the cases aren’t resolved and millions of dollars have been incorrectly collected. <strong>It would be easy to criticize this issue as a “bad apple” that fell from the AI tree but as automated systems are poised to become the spearhead of modernization initiatives, cases like this one cannot be classified as isolated incidents.</strong></p><p>Even if that case seems a bit extreme, it should act as a cautionary tale. <strong>As more and more organizations rely on AI to modernize critical systems, a new inevitable challenge arises: How can we trust such systems</strong>? Companies are starting to grasp the reality of relying on predictive applications for their business, it is only truly valuable when every stakeholder understands how decisions are made.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/800/0*7Gw-jeRfujLlTTSr.jpg" /></figure><h3>Introducing Explainable AI (XAI)</h3><p>When thinking of automation, the best way to avoid bad surprises is to approach it from a risk management perspective. In other words, <strong>we need to think how to design AI processes with human supervision in the loop</strong>. Let’s go through an example. Imagine you own a bank (dream big). In order to stay competitive you decide to implement an automated system that grants or denies loans based on your applicants’ data. On paper, it’s win-win scenario. Customers get an answer in minutes when it used to take days. Time and resources allocated to the task are reduced significantly and your bottom line is positively impacted. Congratulations? … Not so fast! It is important to remember that AI systems are not standalone applications. In production environments, they are part of an ecosystem and need to cooperate with the other actors. <strong>The difference with AI is that you are introducing a system that creates its own rules in an ecosystem that traditionally follows a pre-defined playbook</strong>. If you cannot clearly understand the rules by which your models play, <em>aka a black-box AI</em>, clashes will be inevitable at every level of your organizations. No matter how well your AI system is performing, risks are unmanageable by design if it is a black-box for the people whose job is to “cooperate” with it.</p><p>In our “banking” scenario, a black-box AI can create problems ranging from a frustrating user experience to company-wide compliance issues. Let’s illustrate it with examples on how the different stakeholders can react to your system:</p><blockquote><em>Customer: why is my loan application denied?</em></blockquote><blockquote><em>Customer support: how do I answer this customer complaint?</em></blockquote><blockquote><em>IT operations: How do I monitor and debug this model?</em></blockquote><blockquote><em>Data Scientists: Is this the best model I can build?</em></blockquote><blockquote><em>Internal Audit / Regulators: Are these credit scoring models fair?</em></blockquote><p><strong>Your ability to give clear and transparent answers at every stage of the process will determine the usefulness of your models.</strong> For this particular reason the field of Explainable AI (XAI) aims to provide a comprehensive framework for AI practioners to future-proof their models. We will keep the technical recommendation for another article and will focus here on the things to consider to democratize explainable AI in your organization.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/681/0*EFAW6pQ5mrvN7RBL.png" /><figcaption>What is XAI? — <a href="https://www.kdnuggets.com/2018/10/enterprise-explainable-ai.html">KDnuggets</a></figcaption></figure><h3>XAI and data teams</h3><p>As models grow more complex, interpretability becomes a challenge. Depending on what algorithms are used, it is possible that no one, including the algorithm’s creators, can easily explain why the model generated the results that it did. <strong>Choosing the right model is about navigating the trade-offs between accuracy and explainability</strong>. For example, decision trees are easy to interpret by design, their predictions resulting from a chain of easily traceable sequential steps. But on the other end of the spectrum you have neural networks which provide a high predictive accuracy in cases such as computer vision (great for unstructured data types like images or documents), but at the expense of interpretability.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*83O5WqvC16p_wgjX.png" /><figcaption>Slide from “<a href="https://xaitutorial2020.github.io/">AAAI 2020 Tutorial on Explainable AI</a>”</figcaption></figure><p>Today, <strong>the choice of algorithms is also constrained by the regulatory frameworks in place</strong>. For example, the <a href="https://journals.sagepub.com/doi/abs/10.2501/IJMR-2017-050?journalCode=mrea">European GDPR regulation</a> states that “the existence of automated decision-making, should carry meaningful information about the logic involved, as well as the significance and the envisaged consequences of such processing for the data subject”. A challenge that covers more than just selecting the right models. Bias in datasets and features are model agnostic but their impact on your model cannot be overlooked. <strong>Your AI systems will only be as good as the examples they can learn from</strong>. If your lending app regularly discards loan applications from minorities (<a href="https://www.bankingdive.com/news/artificial-intelligence-lending-bias-model-regulation-liability/561085/">a known issue</a>) you need to be able to fix it quickly.</p><blockquote><strong><em>This type of problem are usually the result of data teams living in a bubble. Most organizations structures their data science operations in silos. Separation from the domain expert teams deprives them from a valuable feedback loop. Time for a little introspection, who owns AI initiatives in your company?</em></strong></blockquote><h3>Enterprise XAI: beyond the data team</h3><p>Companies with AI experience were the first to understand that a full stack data scientist does not exist, therefore collaboration with other parts of the business is what will make or break any AI strategy.</p><blockquote>The people who raised the ethics issues with banks — the original ones — were the legal and compliance team, not the technologists. The technologists want to push the boundaries; they want to do what they’re really, really good at. But they don’t always think of the inadvertent consequences of what they’re doing.” — <strong>Paul Moxon, SVP Data Architecture at Denodo Technologies — </strong><a href="https://www.nextgov.com/emerging-tech/2020/06/who-should-lead-ai-development-data-scientists-or-domain-experts/166119/"><strong>Nextgov</strong></a></blockquote><p>But collaboration with members outside of the data team is the struggle today. The lack of a shared language between both sides is to blame. <a href="https://venturebeat.com/2019/07/19/why-do-87-of-data-science-projects-never-make-it-into-production/"><strong>87% of AI projects never make it to production</strong></a><strong> and </strong><a href="https://blogs.gartner.com/andrew_white/2019/01/03/our-top-data-and-analytics-predicts-for-2019/"><strong>80% resemble alchemy</strong></a><strong>, run by wizards whose talents will not scale within their organization</strong>. As the AI hype is fading (no, robots won’t replace you) companies are realizing that throwing money at the problem and hiring more data scientists won’t magically create value, especially in global context of AI skills shortage. The focus is now shifting to the training of domain experts and how they can gain AI skills. Instead of PhDs, what about training domain experts on how to utilize existing AI tools? <strong>No need to reinvent the wheel, we need better drivers.</strong></p><p>In <a href="https://www.cray.com/resources/enterprise-ai-adoption-survey">The State of Enterprise AI Adoption (Cray — 2019)</a>, 95% of respondents stated that AI will become critical to their business capability sometime over the next 3 years and the lack of technical expertise was the biggest challenge ahead, second to the cost of deploying AI systems. While most companies understand the opportunities lying ahead, <strong>the market is still in the “education phase” of the AI adoption lifecycle, trying to pin down the use cases that will create most value</strong>. The process is slow because educating the workforce on AI is a challenge in itself. The learning curve is very steep and most are beginners with little to no background to follow theory/code-based courses. AI training content for domain experts is still lacking but the need is real. In the same survey, <strong>72% of respondents said they have participated in one or more activities to educate themselves on AI in the past year.</strong></p><p>It’s safe to say that the future of AI will follow the traditional technology adoption curve: it’s only when non-experts can create value with it that AI will truly become a commodity!</p><blockquote>(As opposed to a Data Scientist) an informed business owner understands the domain, the context and the consequences — the true consequences — of the wrong answer. — <strong>Richard Eng, Chief Engineer at The MITRE Corporation — </strong><a href="https://www.nextgov.com/emerging-tech/2020/06/who-should-lead-ai-development-data-scientists-or-domain-experts/166119/"><strong>Nextgov</strong></a></blockquote><h3>Deepflow: AI collaboration platform for teams</h3><h3>&gt;&gt;<a href="https://bit.ly/Deepflow-Early-Adopter">Join the next batch of early adopters</a>&lt;&lt;</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*w0MnylOucp9G19iUG3Ap9Q.jpeg" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1285cb3e9fe8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/deepflow-ai/designing-trustworthy-ai-applications-1285cb3e9fe8">Designing trustworthy AI applications</a> was originally published in <a href="https://medium.com/deepflow-ai">Deepflow AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Introducing Flows: from data to AI without a single line of code]]></title>
            <link>https://medium.com/deepflow-ai/introducing-flows-from-data-to-ai-without-a-single-line-of-code-70f1126a5f84?source=rss----eeca9931d9b---4</link>
            <guid isPermaLink="false">https://medium.com/p/70f1126a5f84</guid>
            <category><![CDATA[no-code]]></category>
            <category><![CDATA[low-code]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[data-science]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Vincent Van Steenbergen]]></dc:creator>
            <pubDate>Mon, 18 May 2020 14:03:10 GMT</pubDate>
            <atom:updated>2020-06-19T08:22:31.993Z</atom:updated>
            <content:encoded><![CDATA[<p><em>This post was </em><a href="https://www.deepflow.ai/introducing-flows-from-data-to-ai-without-a-single-line-of-code/"><em>first published on our blog</em></a><em>.</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fwbcTbGOtXGDK-tg0scSVQ.jpeg" /></figure><p><em>In a </em><a href="https://www.deepflow.ai/putting-data-science-in-the-hands-of-domain-experts/"><em>previous post</em></a><em> we looked at why enterprise AI is broken and introduced Deepflow, our solution to put data science in the hands of non-developers.</em></p><p>Traditionally, there are two ways your organization can adopt AI at scale. You can go the Horizontal or the Vertical route. Horizontal AI tools are made to be flexible and use case agnostic. They are development platforms that give you full control over your data science workflows, from conception to deployment. On the other side you have Vertical AI tools. Made for specific use cases, these tools are designed to solve one problem and one problem only. They do it well and the learning curve is often little to non-existent.</p><p>Once an organization sees the value in using already existing AI tools instead of reinventing the wheel, the choice between adopting Horizontal or Vertical tools will fall on who owns AI initiatives internally. Data scientists and engineers are accustomed to having flexible tools with great levels of customization. If they choose on behalf of everybody else, they will naturally push for implementing Horizontal tools, de facto excluding anyone not on the technical side. Horizontal solutions, by wanting to be a one-stop-shop for everything AI, create a new set of problems. Customizing them quickly turns into maintaining expensive <a href="http://softwarecreation.org/2008/how-a-beautiful-software-system-becomes-frankenstein/">Frankenstein​ softwares</a> and it makes it even harder for your team to move from research to production. So what about Vertical tools then? Well, their ease-of-use cannot exist without a lack of transparency. Most of them suffer from a black box approach, forcing you to trust the outputs blindly. They are often hard or impossible to customize and their pricing models makes their cost at scale increase exponentially.</p><p>It should be clear by now that the path to enterprise AI is a catch 22 for most companies. Favoring accessible solutions makes you lose freedom while flexible solutions lock non-technical employees out. The idea behind Deepflow stems from this conundrum. We decided to combine the level of customization offered by Horizontal solutions with the ease-of-use unique to Vertical tools. And to achieve this we had to rethink how people approach working with data. That is why Deepflow is built around the concept of <strong><em>Flows</em></strong>, a brand new way to experience data science by approaching every aspect of working with data through a drag and drop graphical interface.</p><h3>Go with the Flow!</h3><p>In Deepflow, a <strong><em>Flow</em></strong> is what we call an end-to-end data science workflow. But that is just scratching the surface. <strong><em>Flows</em></strong> are much more powerful than that. <strong><em>Flows</em></strong> are designed to be:</p><ul><li>Visual</li><li>Collaborative</li><li>Always production-ready</li></ul><h4>Flows are Visual</h4><p>Work with data visually from start to finish. <strong><em>Flows</em></strong> are designed to get you from data to AI quickly and without a single line of code. Their visual nature is the antidote to AI’s black box problem. <strong><em>Flows</em></strong> are transparent and auditable by design. Spend time thinking about high level data problems, rather than mindlessly repeating busy work.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/730/0*lVdaj_ZKwHPIu6cx.png" /></figure><ul><li><em>Drag &amp; drop interface made for humans</em></li><li><em>Automate end-to-end data work visually</em></li><li><em>Build trust in AI with transparent processes</em></li></ul><h4>Flows are collaborative</h4><p>Contribute together as a team in real-time. <strong><em>Flows</em></strong> empower everyone in your team with a shared visual language. Domain experts now have the possibility to design and prototype in tandem without engineers in the loop. Bring your ideas to life faster and test concepts earlier and more often. Show the world, don’t tell.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/730/0*tT273wZ-oClrZcpc.png" /></figure><ul><li><em>Edit and comment alongside your co-workers in real-time</em></li><li><em>Allow non-developers to create value with data</em></li><li><em>Stay on track with auto-save and version history</em></li></ul><h4>Flows are always production-ready</h4><p>Deploy in 1 click and iterate faster. <strong><em>Flows</em></strong> are designed to bring your projects from concept to production even faster. They are always 1 click away from being deployed and they scale automatically. Everyone in your team can easily reuse components and iterate from previous projects. Do not reinvent the wheel!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*cRIb5QbRRpFe8XsU.png" /></figure><ul><li><em>All-in-one DataOps</em></li><li><em>Manage team members with user permissions</em></li><li><em>Endless possibilities with Plugins &amp; API</em></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fwbcTbGOtXGDK-tg0scSVQ.jpeg" /></figure><h3>Register now to join the next batch of early-adopters</h3><p><a href="https://bit.ly/Deepflow-Early-Adopter">&gt;&gt;REGISTER HERE&lt;&lt;</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=70f1126a5f84" width="1" height="1" alt=""><hr><p><a href="https://medium.com/deepflow-ai/introducing-flows-from-data-to-ai-without-a-single-line-of-code-70f1126a5f84">Introducing Flows: from data to AI without a single line of code</a> was originally published in <a href="https://medium.com/deepflow-ai">Deepflow AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Putting data science in the hands of domain experts]]></title>
            <link>https://medium.com/deepflow-ai/putting-data-science-in-the-hands-of-domain-experts-1bdd090d860d?source=rss----eeca9931d9b---4</link>
            <guid isPermaLink="false">https://medium.com/p/1bdd090d860d</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[data-science]]></category>
            <dc:creator><![CDATA[Vincent Van Steenbergen]]></dc:creator>
            <pubDate>Mon, 27 Apr 2020 14:40:30 GMT</pubDate>
            <atom:updated>2020-06-19T08:21:41.956Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Introducing Deepflow, the 1st low-code data science platform. No coding or IT required.</em></p><p>Working with data always feels like being in the trenches. Thankfully, today’s technological landscape bares almost no resemblance to what us, the data scientists of the stone age, had to work with. Looking back at my first machine learning (ML) application, there was as much coding as there was finger-crossing and self-inflicted pep talks. But 10 years after deploying my first ML algorithm in production I still feel like the learning curve for implementing common use cases is too steep for beginners.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/387/1*XFqP9huH_n7J_SYlDpDQ8w.png" /><figcaption>Common AI strategy</figcaption></figure><p>During these years, I’ve had the opportunity to design and implement various machine learning applications ranging from pharmacology clinical trials to recommender systems in retail as well as fraud detection in telecom and finance.</p><p>After witnessing <a href="https://conferences.oreilly.com/strata/strata-eu-2017/public/schedule/detail/57391">the good, the bad and the ugly</a> of what AI has to offer, I realized that while developer tools have continuously improved, more accessible solutions have not been prioritized. And it is not without consequences.</p><p>Today, a widespread lack in accessible data science tools often leaves domain experts out of the loop, resulting in data scientists and engineers setting the pace for AI adoption in their organizations. And it is causing problems. Even with fancy PR doing a great job at praising the miracles of enterprise AI, reality remains far less glorious, <a href="https://venturebeat.com/2019/07/19/why-do-87-of-data-science-projects-never-make-it-into-production/">87% of AI projects never make it to production</a>. It’s an open secret among data practitioners. Afraid to miss the AI train to the benefit of their competitors, too many companies keep throwing resources at the problem in hope for breakthroughs to emerge. Lack of business-AI alignment is poised to set failed AI strategies on repeat.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/817/1*4TnAs5myeYCBMoV3acvZwA.png" /><figcaption>Aim for the moon. Fail. Repeat.</figcaption></figure><p>Enterprise AI is on a collision course with reality and it is only a matter of time before companies wake up to the fact that <a href="https://blogs.gartner.com/andrew_white/2019/01/03/our-top-data-and-analytics-predicts-for-2019/">80% of their AI projects resemble alchemy</a>, run by wizards whose talents will not scale within their organization.</p><p><strong>Tomorrow’s leading companies will be the ones able to put data science in the hands of ALL their employees.</strong> Empowering domain experts is the best strategy to make sure organizations steer away from what Cassie Kozyrkow, Chief Decision Scientist at Google, calls “<a href="https://towardsdatascience.com/artificial-intelligence-do-stupid-things-faster-with-more-energy-379aa6bac220">Machine Learning: doing stupid things faster with more energy</a>”.</p><p>That is why we built <strong>Deepflow</strong>, the first low-code platform to design and run data science workflows 100% visually. No coding or IT required. With our unique approach we strive to future proof companies by making sure AI capabilities are spread evenly between employees, no matter their technical skills. As the data science community grew, we replaced Java by Python and Weka by Tensorflow, now it is time for data science to embrace its “spreadsheet revolution” and truly become a commodity.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fwbcTbGOtXGDK-tg0scSVQ.jpeg" /></figure><h3>Register now to join the next batch of early-adopters</h3><p><a href="https://bit.ly/Deepflow-Early-Adopter">&gt;&gt;REGISTER HERE&lt;&lt;</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1bdd090d860d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/deepflow-ai/putting-data-science-in-the-hands-of-domain-experts-1bdd090d860d">Putting data science in the hands of domain experts</a> was originally published in <a href="https://medium.com/deepflow-ai">Deepflow AI</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>