David DeSanto, CEO of Anaconda – Interview Series

David DeSanto is Chief Executive Officer at Anaconda, where he leads the company’s mission to empower the world’s data science and AI communities through open-source innovation and secure enterprise solutions. A proven product and technology executive, David brings more than two decades of experience spanning cybersecurity, developer platforms, and enterprise software.

Most recently, David served as Chief Product Officer at GitLab (GTLB ), where he led the global product organization in delivering a comprehensive, AI-native DevSecOps platform with more than 50 million registered users worldwide. During his six years with the company, he helped transform GitLab from a high-growth startup into a publicly traded, industry-defining leader of the DevOps Platform category.

Anaconda is a leading open-source platform for data science, machine learning, and artificial intelligence, built around the Python programming language and widely used by both individual developers and large enterprises. Originally launched in 2012, it provides a comprehensive environment that includes tools for coding, package management through Conda, and access to thousands of pre-built libraries such as NumPy, pandas, and TensorFlow, enabling users to develop, test, and deploy AI models efficiently.

Over time, Anaconda has evolved into a full enterprise AI platform that helps organizations manage the entire AI lifecycle—from sourcing and securing open-source packages to building, governing, and deploying applications across cloud and on-premise environments. With tens of millions of users and adoption across a large percentage of Fortune 500 companies, it has become a foundational layer for modern AI development, emphasizing open-source innovation, scalability, and secure, reproducible workflows.

You spent nearly six years at GitLab, over three as their Chief Product Officer, helping scale an AI-native DevSecOps platform to tens of millions of users. How has that experience shaped your priorities now as CEO of Anaconda, and what feels fundamentally different about leading a company versus leading product?

My time at GitLab really reinforced a few principles that are now central to how I’m approaching Anaconda. First is responsible growth—scaling teams, products, and revenue in a way that’s durable. At GitLab, we grew to serve tens of millions of users, and seeing GitLab Ultimate account for over half the company’s revenue demonstrated how important it is to align product value with long-term business impact.

Second is a mindset of results and efficiency over process and structure. It’s okay to ship something that’s good enough and shows direction to get the flywheel of customer feedback going. Delivering real value quickly is imperative, but you still need to be thoughtful about how you scale. That ties closely to the third pillar: customer obsession and really meeting users where they are. I’ve spent my career building developer and security tools, and as a former developer myself, I know how much good (or bad) tooling can impact productivity and satisfaction.

And finally, transparency with purpose. This core value of Anaconda has allowed for all parties to participate and collaborate in making the company and its offerings better. I look forward to further building on this value to ensure we’re giving our community what they need to be successful.

As CEO, you’re responsible for the entire system—strategy, culture, operations, and outcomes. I’m still deeply connected to the product, but I’m thinking broader and longer term. I have to ensure the company is growing responsibly, supporting its people, and delivering value to our customers across every dimension of the business. These are the principles I aim to keep building on here.

What motivated you personally to step into the CEO role at Anaconda, and what convinced you that this was the right platform to build the next chapter of enterprise AI?

I tell everyone there were four reasons I was excited to join Anaconda as CEO. First, the technology. As a developer, I’ve been familiar with and have used Anaconda for a long time. I know how powerful it is and can be. The platform already enables so much, and the foundation the team has built will give us the opportunity to shape what comes next in the AI-native era!

Second, the community. I believe deeply in the power of the open source community. Very few companies have a community as broad and engaged as Anaconda’s.

Third, the people. It’s rare to find this level of leadership in one place. The executive team is exceptional, and their passion is real. We’re building the future with open source and AI, and being part of something this meaningful and impactful has us all genuinely excited.

And finally, the opportunity. This is what ultimately made the decision for me. Anaconda sits at the center of making AI more accessible, helping enterprises build, secure, deploy, and monitor AI at scale. When you combine world-class technology, a vibrant community, and a team like this, you get a rare chance to shape how AI and data science are built and used. That’s what drew me in.

Open source powers the majority of modern AI development, yet many enterprises still struggle to trust it at scale. Why do you believe open source remains the most powerful foundation for AI, and where do you think it’s most misunderstood?

There is a common misconception that the most secure code is code that is hidden as only select people can see it. This is like an ostrich sticking its head into the sand to hide itself. Open source software is the opposite of that. Open source is transparent, welcomes everyone to contribute, and gives organizations around the world a higher set of eyes ensuring the code is secure and operates as expected.

We have not seen a technology rapidly mature as AI has been maturing. For AI to continue its acceleration, you need modern code that moves as fast as possible. Open source does this which is why it is the foundation for modern AI development.

At Anaconda, we lean into that. Our core capabilities and Python ecosystem are open source because that’s the best way for teams to get started and innovate quickly. On top of that, we layer enterprise-grade capabilities, giving organizations the governance, security, and reliability they need to use open source at scale.

AI failure rates in the enterprise remain high, especially with generative AI pilots. From your perspective, what are the core reasons these initiatives stall, and how can infrastructure choices make or break long-term success?

Many organizations have run pilots. Some have solid proof-of-concept projects, and others have a handful of internally-built tools that genuinely save teams time. But very few have moved AI into real production that runs across the entire business. There’s a big difference between “we’re experimenting” and “this is how we work now.” That gap is where most companies are stuck – and it’s not because the technology doesn’t work.

The demo almost always looks good, but the problem shows up when you try to reproduce the demo at enterprise scale. Suddenly, you’re dealing with data governance questions, security concerns, reliability issues, and a fundamental trust problem: Will this application work reliably and keep our data secure? Those issues don’t show up in the demo and thus become an afterthought for companies.

The barrier to AI success isn’t capability, but infrastructure and process maturity. The organizations pulling ahead are led by those choosing to invest in modern foundations where trust and velocity coexist and where built-in security and governance accelerate rather than obstruct. Fragmented toolchains and environments force you to choose between the two, but modern, unified infrastructure and modern AI processes let you have both. Today, removing bottlenecks is creating your competitive edge. This is not just a technical ambition; it’s a business imperative to compete and survive in today’s market. Success will be increased by those who invest in security and governance from the foundation.

You’ve led teams across cybersecurity, product, and developer platforms. How are you bringing that security-first mindset into Anaconda’s strategy around dependency management, environment reproducibility, and supply chain risk?

My path into security started in healthcare, where I saw exactly what scarably passed for ‘secure’ at the time. I got fully immersed into security and it became a passion. As AI workloads, models, agents and integrations become more complex, security risks are multiplied faster than governance can keep up. And even when security risk is managed, the environment itself becomes an obstacle.

Security and compliance with AI are challenging, especially getting real visibility into risk across development and production. That’s where we’re focused. We’re building deeper security capabilities within environment management, additional governance around AI packages outside of the Python ecosystem, and helping reduce risk with AI models through security scanning of their posture. The goal is simple: help organizations move faster with AI while maintaining the visibility, privacy, and resilience they need to trust it at scale.

There’s growing skepticism around AI ROI, alongside a surge in experimentation and vibe coding. How do you distinguish between productive experimentation and enterprise-ready AI systems that actually deliver measurable value?

This year may actually be the first time that AI’s ROI is measured well. Everyone in the industry is treating “efficiency” as time saved, however that isn’t the best topline KPI. Organizations that take the time to create bespoke KPIs directly tied to what is the most important to them will have better results. This can be reduced time in code review for your development team or the quality of lead generation for your go-to-market team. Time by itself and measuring token consumption do not directly signal efficiency.

Anaconda sits at the center of Python-based AI development. How do you see the role of Python environments evolving as organizations move from experimentation to fully governed, production-scale AI systems?

Python is the premier language for AI, and while I don’t expect that to change any time soon, languages always ebb and flow in popularity. Organizations need tools that evolve and natively address underlying performance and scalability issues, enabling AI agents to successfully create enterprise-grade applications and services. I expect organizations will start investing in universal building blocks that accelerate AI value and adoption. That’s what will best position them to navigate the ever-evolving lexicon of coding languages that power AI’s infrastructure layer.

You’ve worked closely with regulated industries and security-conscious enterprises in the past. What does enterprise-grade AI governance really look like in practice, beyond policy documents and compliance checklists?

Enterprise AI and AI-native applications are something completely different than traditional software development. When you treat AI as traditional development, you end up with broken security and governance which will stall innovation. Enterprise AI needs AI-native development best practices where the AI model is the primary component that drives versioning and everything else around it is secondary.

AI governance is the difference between scaling successfully and stalling out. Enterprise-grade AI governance is what happens when responsible AI-native principles are translated into enforceable platform controls, clear accountability, and continuous traceability and lineage of all AI components. It goes beyond policies and checklists that worked in DevOps practices.

With Anaconda’s recent funding and enterprise push, what are your near-term growth priorities, and where are you investing most aggressively over the next 12 to 18 months?

Our goal is clear: Anaconda will be the preferred enterprise AI-native development platform for building, securing, and deploying AI-native applications and services. Our customers continue to morph to meet their needs and we morph with them. This is why we aren’t just a data science toolkit anymore, but a comprehensive platform for enterprise AI. Every investment, whether in product, partnerships, or M&A, will be evaluated against a single question: “does this move us closer to being the platform enterprises need to be successful in the new AI-native world?” We are a customer obsessed company and everything we do is for our customers.

As someone who has helped scale a publicly traded developer platform, what lessons have you learned about building for practitioners while also serving CTOs, CIOs, and Chief AI Officers at the executive level?

Success always comes back to the customer and supporting the outcomes they’re trying to achieve. It sounds simple, but it’s easy for teams to get caught up in internal metrics (how many hours went into a project, whether a launch hits predefined goals) rather than asking a more important question: Does this actually make our customers more successful? Leaders may say a net-new product is needed when in reality, it could just be refining what you already have (making a product more user-friendly, for example), and then you get the outcome you’ve been looking for. If the customers are happy and successful, then we’re all happy and successful.

Thank you for the great interview, readers who wish to learn more should visit Anaconda.