Episode

Humility in the Age of Agentic Coding

Podcast: Practical AI
Published: Mar 17, 2026
Duration seconds: 3326
Processing state: processed
Canonical source: https://share.transistor.fm/s/7e1ca2c8
Audio: https://pscrb.fm/rss/p/dts.podtrac.com/redirect.mp3/media.transistor.fm/7e1ca2c8/3ade85d6.mp3
JSON: /v1/public/podcasts/practical-ai/episodes/humility-in-the-age-of-agentic-coding
Markdown: /podcast/practical-ai/humility-in-the-age-of-agentic-coding.md

Actions

POST https://stenobird.com/v1/public/podcasts/practical-ai/episodes/humility-in-the-age-of-agentic-coding/transcription-requests
Idempotently request low-priority transcript generation for this episode.
GET https://stenobird.com/podcast/practical-ai/humility-in-the-age-of-agentic-coding.md
Read the agent-friendly Markdown representation of this episode resource.

Summary

Software engineer Steve Klabnik shares his transition from AI skeptic to active experimenter using agentic tools. He discusses building the Rue programming language and the tension between rapid development velocity and maintaining code quality.

Topics

Agentic Coding
Software Engineering
Rust Programming Language
AI Productivity
Code Quality
Programming Language Design
Claude AI
Developer Workflow

Highlights

Main idea: AI agents act as a powerful force multiplier for individual developers, enabling rapid prototyping of complex projects like Rue
Failure mode: The temptation of high velocity can lead to 'blind merging,' where developers bypass thorough code reviews, risking long-term technical debt
Practical takeaway: AI tools are most effective at overcoming the 'blank page' problem, transforming the developer's role from writer to editor
Main idea: Increased developer productivity does not necessarily lead to fewer jobs, but rather an infinite expansion of the possible backlog
Critical challenge: The industry must find a way to implement automated validation and trust mechanisms to maintain software quality in an era of agentic acceleration

Chapters

1:00 Introduction: Hosts introduce guest Steve Klabnik and discuss his background in the Rust ecosystem.
5:00 From Skeptic to Experimenter: Steve discusses his evolving perspective on AI and how his views shifted through hands-on use.
13:25 Overcoming the Blank Page: The utility of AI in generating initial drafts and the shift from writing to editing.
26:10 The Future of Engineering Teams: A debate on whether AI-driven productivity will lead to industry layoffs or expanded project scopes.
34:35 Building Rue with Agents: Insights into the development of the Rue programming language using Claude and agentic workflows.
47:00 The Quality vs. Velocity Dilemma: The danger of losing oversight when AI tools allow for high-frequency, low-scrutiny code commits.
51:05 Closing Thoughts: Reflections on the future of software development and the need for trust in automated tools.