Building Claude Code with Boris Cherny - The Pragmatic Engineer Recap
Podcast: The Pragmatic Engineer
Published: 2026-03-04
Duration: 1 hr 37 min
Summary
In this episode, Boris Cherney discusses the evolution of Claude Code at Anthropic, his unconventional entry into software, and the transformative impact of AI on engineering workflows.
What Happened
Boris Cherney shares his unique journey into the tech world, which began at a young age when he started selling Pokémon cards on eBay. His curiosity led him to HTML, where he learned to enhance his listings, and later to programming on TI-83 calculators to ace math tests. This practical approach to coding ultimately transitioned him into a series of startups, where he navigated the unpredictable landscape of entrepreneurship and product development. His experiences laid a foundation for his role as engineering lead at Anthropic, where he now leads Claude Code, a rapidly advancing developer tool.
At Anthropic, Boris faced an unusual challenge: his first pull request was rejected, not due to poor code, but because it was handwritten rather than generated by AI. This incident highlights the shift in coding practices within the company, where AI-generated code is becoming the norm. Boris discusses his workflow of shipping 20-30 pull requests daily with zero handwritten code, emphasizing the efficiency and speed that AI brings to software development. He also dives into the internal discussions at Anthropic regarding the potential release of Claude Code, weighing its implications for the developer community and the future of AI in engineering.
Boris reflects on the transformative nature of our current era, comparing it to the advent of the printing press. He stresses that some engineering skills are becoming less relevant as AI tools evolve, while others are gaining importance. This shift poses questions for engineers about how to adapt and thrive in a landscape defined by rapid technological change. This episode is a deep dive into the intersection of AI and software engineering, offering valuable insights for anyone interested in the future of coding.
Key Insights
- Boris Cherney's unconventional path to software engineering began with practical applications of coding in his youth.
- Claude Code's development reflects a significant shift towards AI-generated coding practices in modern engineering.
- The rapid evolution of AI tools is comparable to historical technological transformations like the printing press.
- Engineers must adapt by honing relevant skills as traditional coding practices become less significant.
Key Questions Answered
What was Boris Cherney's early introduction to coding?
Boris Cherney's journey into coding began when he was around 13 years old, selling his old Pokémon cards on eBay. He discovered he could write HTML to enhance his listings, and he found that using the blink tag allowed him to sell cards for higher prices. This sparked his interest in coding, leading him to learn HTML further. Additionally, he programmed answers into his TI-83 graphing calculator for math tests, progressing from basic programming to assembly language as the math became more complex.
How did Claude Code evolve at Anthropic?
Claude Code began as a side project and quickly gained traction as one of the fastest-growing developer tools. Boris discusses the internal debate at Anthropic regarding whether to release Claude Code, showcasing the challenges and considerations faced by the team. The project's evolution reflects a broader trend in software development, where AI-generated code is increasingly common, fundamentally changing how engineers approach their work.
What challenges did Boris face when he joined Anthropic?
Upon joining Anthropic, Boris faced an unexpected challenge: his first pull request was rejected not due to issues with the code itself, but because it was handwritten. This incident highlights the cultural shift within the company, where AI-generated code has become the standard. Boris's experience underscores the rapid changes in coding practices and the need for engineers to adapt to new technologies.
What does Boris believe about the current era of engineering?
Boris believes we are living through a transformative time in engineering, akin to the impact of the printing press. He argues that the evolution of AI tools is reshaping the skills that engineers need, as traditional coding practices become less relevant. This perspective encourages engineers to rethink their roles and focus on developing skills that align with the future of software development.
How did Boris's startup experiences shape his approach to product development?
Boris's experiences in various startups taught him the importance of adaptability and the need to pivot based on market demands. He emphasizes that startup success is rarely linear; it requires forming hypotheses and being willing to change direction based on user feedback. His journey highlights the necessity of finding product-market fit and the surprises that come along the way, which are critical lessons for anyone involved in product development.