▲ Community Session: Vercel 🖤 Python

Highlights

• Vercel has acquired GelData to significantly bolster its Python capabilities and become a leading "AI cloud."

• Yuri Selivanov and Elvis Pranskevicius

Timeline

1. 00:00 - 05:06

   Introductions and Speaker Backgrounds

   The session opens with hosts Amy Egan and Jacob introducing guests Elvis Pranskevicius and Yuri Selivanov following Vercel's acquisition of GelData. Elvis describes his background creating asyncpg, the de facto asynchronous driver for Postgres, while Yuri recounts his journey as a Python Core developer since 2013. Yuri explains his work on async/await syntax and the creation of UVloop, which serves as a high-performance drop-in replacement for the standard Python event loop. This section establishes the speakers' deep technical authority within the community and sets the stage for Vercel's ambitious goal to become the world's best Python cloud. Their transition marks a shift from building databases to improving the foundational Python experience at Vercel.

2. 05:07 - 09:48

   Why Vercel is Investing in Python Now

   The discussion shifts to why Python has become a strategic priority for Vercel at this specific moment in tech history. Yuri explains that while Python was always growing, the current AI revolution led by labs like Meta and OpenAI has pushed it into a new cycle of extreme relevance. He contrasts the mature developer experience (DX) Vercel offers for JavaScript with the still-difficult process of managing and scaling Python applications in production. Elvis adds that Vercel is evolving from a "frontend cloud" to an "AI cloud," where Python's role in backend logic and data processing is indispensable. They argue that the time is right for Vercel to replicate the "magic" of its JavaScript tooling for the Python ecosystem.

3. 09:49 - 15:13

   Bridging the Gap Between Python and JavaScript Tooling

   Yuri shares a fascinating historical anecdote about a missed opportunity where Python almost became the primary browser language instead of JavaScript. He explains that because this didn't happen, Python's tooling evolved to focus on backend pipelines and machine learning rather than UI transpilers and CSS modules. The goal at Vercel is not to invent new Python frameworks, but to create tools that seamlessly connect existing Python projects to Vercel's elastic infrastructure. Elvis introduces the concept of "framework-defined infrastructure," where the platform automatically configures itself based on the user's code. This approach ensures that developers can use familiar tools like Git and the Vercel CLI while gaining observability and scalability for free.

4. 15:14 - 20:15

   The Strength of the Python Ecosystem

   The speakers address the specific strengths that make Python a preferred choice for developers, famously describing it as the "second best language for everything." They highlight the immense depth of the PyPI package repository and the language's long-standing dominance in academia, finance, and scientific computing. Yuri notes that Python acts as the perfect "glue language," orchestrating high-performance low-level code written in C++ or Rust for GPU-intensive AI tasks. While some developers prefer a pure TypeScript stack for simplicity, Yuri argues that a diverse tech stack often leads to more inspiration and better engineering outcomes. The community chat echoes these sentiments, noting that Python's love story with machine learning and data pipelines is unmatched by any other language.

5. 20:16 - 28:44

   Advice for Learning Python and the Role of AI

   When asked for advice for newcomers, the guests strongly advocate for project-based learning over simple tutorials. Yuri and Elvis warn against "vibe coding," where developers ask LLMs like ChatGPT or Cursor to write entire blocks of code without understanding the underlying logic. They suggest asking AI "how" to do something rather than asking it to "do" the work, which ensures the developer builds the necessary neural connections to master the syntax. The speakers also emphasize the value of joining local communities, attending hackathons, and contributing to open-source documentation as ways to gain real-world experience. Yuri points out that software engineering provides a unique feedback loop where a single fix can benefit millions of people almost instantly.

6. 28:45 - 35:36

   Open Source Commitment and Future Hiring

   In the final segment, Yuri confirms that the team will continue and even increase their contributions to Python Core and the standard library while at Vercel. They aim to foster internal talent and mentioned they will eventually look for a Developer Advocate who is passionate about Python and technical blogging. Yuri and Elvis provide their contact information, including Yuri's direct Vercel email, and urge the community to provide honest, even critical, feedback on Python DX issues. They emphasize that while telemetry provides data, it cannot tell a company when documentation is confusing or when a tool feels sub-optimal to use. The session concludes with a reminder that this is the final stream of the year, with more community updates planned for January.