Deepnote CEO Explains Why Notebooks Are the Ideal Interface for AI Agents

When Jakub Jurových and his team began building Deepnote in 2019, they identified a gap between two computational worlds that no existing tool could bridge. “World A was the world of tools that were simple to use, easy to get started with, but they also take you only so far,” said Jurových, Deepnote’s founder and CEO, citing spreadsheets as the prime example. “In World B, where tools are much more advanced, you can build anything that you can imagine — but first you need to spend a lot of time learning the tool.” Jurových set out to build the missing computational middle ground.

The Missing Computational Medium

Jurových appreciated the concept of notebooks, which have existed since the 1980s, but found existing formats weren’t designed for the tight feedback loops data exploration demands. Unlike software engineers with clear tickets and endpoints, data scientists often receive a CSV file and are told to “go find something interesting.” “Data exploration is a completely different way of working,” he explained. “There’s no obvious endpoint, you can always go wider or deeper with data.” Deepnote was designed for constant collaboration rather than asynchronous pull requests. “We showed how there can be not just two or three people pair programming in one notebook, but hundreds of people, all at the same time. And now we are routinely having sessions with thousands of people in one notebook.”

From Scratchpad to Production

The decision to go open source was not immediate. The team wanted to open source Deepnote from day one but prioritized solving stability, reproducibility, and collaboration challenges first. “We realized that it’s important to solve the problems first, and then open source can be just the cherry on top,” Jurových noted. The team also needed confidence in their architecture before committing to backward compatibility. Six years later, Deepnote is ready to go open source with a new format designed for the cloud, collaboration, and the AI era. While Jupyter had two cell types (code and markdown), Deepnote now supports 23 building blocks — and counting. “We see notebooks as a beautiful format where you can actually stay and keep in the same place all the way to productionizing your workflow,” Jurových said. “The notebook itself can become the whole data app. It can become that thing that you schedule to run every 12 hours. It can have an API endpoint attached to it.” This flexible multitasking capacity is why, he concluded, “Notebooks are the perfect user interface for working alongside AI agents.” For more insights, listen to the full episode of The New Stack Makers featuring Jakub Jurových.

---

Source: Originally published at The New Stack on November 24, 2025.

Frequently Asked Questions (FAQ)

Notebooks and AI Agents

Q: Why are notebooks considered the ideal interface for AI agents? A: Notebooks provide a flexible and interactive environment where users can execute code, visualize results, and document their findings in a cohesive manner. This makes them well-suited for collaborating with AI agents, allowing for iterative development, experimentation, and clear communication of complex processes. Q: How does Deepnote facilitate collaboration with AI agents? A: Deepnote is designed for real-time, multi-user collaboration, allowing hundreds or even thousands of people to work together in a single notebook simultaneously. This collaborative environment is crucial for effective teamwork alongside AI agents, enabling seamless sharing of progress and insights. Q: What makes Deepnote's notebook format different from traditional notebooks like Jupyter? A: Deepnote's format has evolved beyond the basic code and markdown cells of traditional notebooks. It now supports 23 different building blocks, with more being added, to accommodate a wider range of computational tasks and integrations, making it more robust for productionizing workflows and creating data applications. Q: Can a notebook developed in Deepnote be used in production? A: Yes, Deepnote aims to allow users to stay within the notebook environment from initial data exploration all the way to productionizing their workflows. A notebook can evolve into a full data app, be scheduled for regular execution, or even have an API endpoint attached to it.

Deepnote's Development and Philosophy

Q: What was the primary motivation behind creating Deepnote? A: The founders identified a gap in existing tools, which were either too simple and limited (like spreadsheets) or too complex and required extensive learning (like specialized software). Deepnote was built to bridge this gap, offering an advanced yet accessible computational environment. Q: Why did Deepnote choose to go open source? A: While initially prioritizing stability, reproducibility, and collaboration, Deepnote is now ready to go open source. This decision is driven by a desire to foster community development and adoption, especially in the AI era.

Crypto Market AI's Take

The convergence of interactive notebooks and AI agents presents a powerful paradigm shift for data science and beyond. Deepnote's CEO highlights how the structured yet flexible nature of notebooks makes them an intuitive playground for AI collaboration. This aligns with our own platform's focus on leveraging AI for market intelligence and trading. At Crypto Market AI, we see the potential for AI agents to analyze market data, identify trends, and even execute sophisticated trading strategies, all within an integrated and accessible environment. The evolution towards more versatile notebook formats like Deepnote's could further democratize access to advanced AI-driven financial tools, making sophisticated market analysis more attainable for a broader audience.