Explores harness engineering concepts to build trust in AI-generated code from coding agents.
Announcing a new podcast episode on Agentic Feature Owners, featuring Jessitron, with reflections on code and life.
Explores which programming language is best for LLM-based code generation, questioning if machine code is viable.
AI reveals most knowledge work is scaffolding overhead, not core tasks, making it easy to automate.
A balanced perspective on using AI and LLMs as tools, not for everything, with practical examples from web browser documentation work.
A balanced perspective on using AI as a tool, discussing when it's useful and wasteful, with examples from web browser documentation work.
Argues spec-driven development with AI tools is waterfall methodology in disguise, advocating for leaner, discovery-based approaches.
Critique of AI's 'frictionless' promise, arguing human expertise and collaboration remain essential despite automation hype.
A daily tech reading list covering AI agents, software engineering practices, LLMs, and developer tools.
Article discusses the true purpose of code review beyond bug-finding, emphasizing judgment, communication, and codebase health.
A developer's month-long experiment with OpenAI's Codex, exploring how AI is fundamentally changing software development and the programmer's role.
Advocates for radical transparency in customer support, sharing detailed technical explanations instead of vague corporate responses.
Introducing Case, a programmatic harness that dispatches AI agents to automate bug fixes and feature development across multiple open-source repositories.
How to use pattern matching and AI agents to scale code review processes in high-output development teams.
Explores how AI agents are evolving from code assistants to autonomous developers, using existing workflows like pull requests for collaboration.
Reflects on the shift in open source from a community-driven ethos of sharing 'lego bricks' to a more commercial, VC-backed model.
A Django core developer warns against over-reliance on LLMs in open-source contributions, emphasizing the need for human understanding.
A Django core developer warns against over-reliance on LLMs in open-source contributions, emphasizing the need for human understanding.
Java 26's 'boring' nature is a strength, focusing on stability, performance, and predictable upgrades rather than flashy new features.
A guide to identifying deceptive 'AI snake-oil salesmen' and their red flags, contrasting them with legitimate AI practitioners.
