The Pragmatic Engineer

11 episodes scanned.

Recent Episodes

1. Software engineering at Google & CS education with Titus Winters

   Oct 29, 2026

   Titus Winters, a former Google engineer who led its readability program and co-authored Software Engineering at Google, discusses how Google's test requirements evolved and how his team carried out a large C++ refactoring across a roughly 250-million-line codebase. The conversation also covers the creation of the Abseil open-source C++ libraries, why traditional productivity metrics miss some of AI's benefits, and his work on the ACM/IEEE undergraduate computer science curriculum. He argues that CS education should shift toward grading process, prompts, and tests rather than only final code, and offers advice for new engineers.

2. The revised rules of engineering management with Will Larson

   Oct 21, 2026

   Will Larson, CTO of Imprint and former engineering executive at Carta, Calm, Stripe, and Uber, discusses how engineering management practices from the 2010s zero-interest-rate era have changed and why. The conversation covers the shifting role of engineers toward solving business problems, the path to CTO, and why line managers may be better positioned than middle managers for engineering executive roles. Larson also shares his early career, including his time at Yahoo and Digg, and his views on writing and visibility for engineers.

3. Uber’s AI infra, from the inside

   Oct 14, 2026

   Uday Kiran Medisetty, a long-tenured engineer at Uber who leads much of its agentic AI work, walks through the internal developer infrastructure Uber built before generative AI, including M3DB, Michelangelo, Cadence, and DevPods. He then covers the newer AI foundations, such as the model gateway with AI Guard, the AIFX harness wrapper, and tooling for background agents and code review. The conversation also touches on token spend management, measuring engineering productivity through feature velocity, and how the engineer's role shifts toward context engineering and verification.

4. Building resilient systems with Sam Newman

   Oct 7, 2026

   Sam Newman, author of Building Microservices and Building Resilient Distributed Systems, discusses his journey from early computer programming through Thoughtworks to independent consulting. He explains why microservices are an "architecture of last resort," shares his three rules for distributed systems, and explores how modular architecture can help teams safely experiment with AI while maintaining system understanding.

5. Distributed databases with Peter Mattis

   Sep 30, 2026

   Peter Mattis, co-founder and CTO of Cockroach Labs, discusses his 30-year career building large-scale distributed systems, from creating GIMP in college to designing Gmail's storage architecture and Google's Colossus distributed file system. He explores how AI coding tools have enabled him to return to hands-on development after stepping back from coding between 2022-2024, allowing him to produce unusually high-quality, production-ready code at scale while amplifying domain expertise.

6. Design Engineering with Maggie Appleton

   Sep 23, 2026

   Maggie Appleton, Staff Research Engineer at GitHub Next, discusses design engineering and the intersection of design, anthropology, and software development. She explores how designers and engineers can learn from each other, the role of AI agents in design workflows, and why pen-and-paper sketching remains essential even as agentic tools transform the design process. The episode covers her journey from cultural anthropology to product design at AI startups like Elicit, her philosophy on designing for developers and complex systems, and her vision for "barefoot developers" who use AI to build local-first software.

7. AI Skills with Matt Pocock

   Sep 17, 2026

   Matt Pocock, a developer-turned-educator and creator of the popular Total TypeScript course, discusses his unconventional path from voice coaching to software engineering and his pivot toward AI skills and strategic programming. He explores how AI has fundamentally shifted what engineers need to learn—from tactical knowledge (now commoditized) to strategic wisdom—and shares his framework for working with AI agents through skills like 'Grill Me' and 'Wayfinder' that encode software engineering best practices into agent workflows.

8. Building Codex with Tibo Sottiaux

   Sep 9, 2026

   Tibo Sottiaux, one of the original engineers who built Codex and now head of Core Products & Platform at OpenAI, discusses the origins and evolution of Codex, the AI coding harness. The conversation covers why Codex was built in Rust and released as open source, how it supports multiple model providers, and the engineering challenges of merging Codex into ChatGPT to serve millions of users at scale.

9. Why performant code matters (but gets widely ignored), with Casey Muratori

   Aug 26, 2026

   Casey Muratori, programmer, game developer, and founder of Molly Rocket, discusses why software performance is widely ignored despite overwhelming evidence of its importance to business outcomes. The conversation covers the historical role of performance in bringing games to Windows, why most developers and companies deprioritize optimization, and why the tide is finally turning with new performance-focused products gaining traction. Muratori emphasizes the importance of understanding theoretical hardware limits, reading assembly language, and proper optimization methodology.

10. From Chrome DevTools to AI Engineering, with Addy Osmani

    Aug 19, 2026

    Addy Osmani, a director of engineering at Google who spent 14 years building Chrome, DevTools, and Core Web Vitals, discusses his path from building a web browser as a teenager in rural Ireland to leading engineering at one of the world's largest tech companies. The episode covers his work on foundational web developer tools, his philosophy of understanding how things work at deeper layers, and his recent transition as he explores the intersection of AI and software engineering, including concepts like cognitive surrender and loop engineering.