More The Pragmatic Engineer episodes

The creator of Clawd: "I ship code I don't read" thumbnail

The creator of Clawd: "I ship code I don't read"

Published 28 Jan 2026

Duration: 6844

Software development practices are being transformed by AI tools that generate code from detailed prompts, requiring a shift to model-driven approaches and careful quality assurance to ensure reliability.

Episode Description

Brought to You By:Statsig The unified platform for flags, analytics, experiments, and more.Sonar The makers of SonarQube, the industry standard for au...

Overview

The podcast explores the transformation of software development in the context of artificial intelligence, with a focus on how AI tools like Clotbot and Codex are reshaping coding practices. The discussion centers on moving from traditional coding to a model-driven approach, in which AI agents generate and refine code based on detailed prompts, reducing the need for conventional pull requests. The speaker emphasizes that while AI can significantly speed up development, it must be carefully integrated and validated to ensure quality and reliability, with a strong focus on testing and closing feedback loops within AI systems.

The conversation also reflects on the challenges of debugging and refining AI-generated code, as well as the evolving role of developers, who may increasingly focus on high-level architecture and strategic prompting rather than low-level coding. The importance of human oversight in areas like code review and documentation is highlighted, underscoring that AI can be a valuable tool but must be guided by a deep understanding of system design. The discussion also touches on the complexity of tasks such as PDF rendering and the need to balance trust in AI-generated outputs with manual refinement to ensure accuracy and performance.

Recent Episodes of The Pragmatic Engineer

29 Oct 2026 Software engineering at Google & CS education with Titus Winters

"Critiques CS education, advocating for AI-informed, practical curricula focused on problem-solving, teamwork, and real-world development, while examining Google's engineering practices, productivity challenges, and the need for deterministic tools in AI-driven software engineering."

21 Oct 2026 The revised rules of engineering management with Will Larson

"Engineering management is evolving, with outdated 2010s practices giving way to business-focused problem-solving, AI-driven role shifts, and adaptable leadership that prioritizes high-ownership teams and reskilling for long-term success."

7 Oct 2026 Building resilient systems with Sam Newman

Recommended: Good Engineering Decisions

"Microservices are a last-resort architecture requiring careful state management, while resilient systems depend on robustness, rebound, extensibility, and adaptability, with AI augmenting but not replacing human oversight in development."

30 Sept 2026 Distributed databases with Peter Mattis

"Software engineer and Cockroach Labs co-founder Peter Mattis discussed his career in database design, distributed systems, AI's impact on development, and the evolution of his leadership role, stressing rigorous testing, security, and the transformative potential of AI in coding."

23 Sept 2026 Design Engineering with Maggie Appleton

"Explores the evolving role of design engineers, blending engineering and design principles, analog tools, AI, and interdisciplinary collaboration to shape intuitive, culturally aware digital experiences."

More The Pragmatic Engineer episodes