- Harness engineering
Birgitta Böckeler on the rise of importance of constraining your development - 96% engineers don’t fully trust AI output, yet only 48% verify it
"With the speed of writing code increasing due to AI coding tools, skills like critical thinking, pure problem-solving, and being good at reviewing/verifying are more important than ever." — Gregor Ojstersek - Stop generating, start thinking
"Where I've seen LLMs do the most damage is where engineers outsource the thinking that should go into software development. LLMs can't reason about what the system architecture because they cannot reason. They do not think. So if we're not thinking and they're not thinking that means nobody is thinking." — Sophie Koonin - How StrongDM’s AI team build serious software without even looking at the code
Simon Willisons commentary on an experiment to build reliable code without review by people - My AI adoption journey
Mitchell Hashimoto shares his journey with adopting AI into his software development workflow - Relocating rigor
"If you're working with generative AI now, the question to ask yourself is: where did the rigor go? ... The engineers who thrive in this environment will be the ones who relocate discipline rather than abandon it." — Chad Fowler - The age of the micro-entrepreneur
"...programming is transforming from syntax-first to systems-first" — Krzysztof Zabłocki - My LLM coding workflow going into 2026
Addy Osmani outlines his current practices - Nine risks caused by AI notetakers
Rachel Coldicutt highlights the challenges that these present in an organisation. - DevAI: Beyond hype and denial
Ivan Kusalic on what works and what doesn't when using AI tools to support development - Vibe engineering
"It’s also become clear to me that LLMs actively reward existing top tier software engineering practices" — Simon Willison - "Human in the loop" is a thought-terminating cliche
Pavel Samsonov makes that case that without thoughtful guard rails, asking people to clean up after machines doesn't make machines better — it makes people worse. - The summer of Johann: prompt injections as far as the eye can see
Simon Willison summarises Johann Rehbergs recent work in publishing security vulnerabilities in ChatGPT, Codex, Anthropic MCPs, Cursor, Amp, Devin, OpenHands, Claude Code, GitHub Copilot and Google Jules. - Writing and thinking with AI: Why repositories beat chatbots
Chris Parsons shares the advantages of using Claude Code over chat interfaces for knowledge work. - What can agents actually do?
Will Larson outlines his definition of AI agents and how to think through what is possible with them. - What I learned from running a series of 1:1 AI clinics with engineering leaders
Yemi Olagbaiye offers some helpful reframings to help you think about AI adoption in your teams - From autocomplete to agents: AI coding assistance state of play
Birgitta Böckeler's talk gives the state of play in June 2025 - My AI skeptic friends are all nuts
Thomas Ptacek makes the case for AI assisted programming - What we talk about when we talk about AI
"Puncturing the glossy, mythical hype bubbles is an important part of understanding AI and making its consequences clear." — Rachel Coldicut - "AI-first" is the new Return To Office
Anil Dashs' reminder of the faddish nature of the bubbles that CTOs and VCs live in. - The implementation paradox: Why your engineering team isn't using AI (and how to fix it)
Toby Moore shares his take on what might be leading engineers in your team to resist using AI. - Thoughts On A Month With Devin
One team's experiences trying Devin for a month - AI for learning: What we learned in 2024
"Start with the problem. Every successful AI initiative we explored identified primarily their challenge and clear outcome" — Matt Walton - How GenAI is reshaping tech hiring
[BEHIND A PAYWALL]
Gergely Orosz and Elin Nilsson provide a deep dive into the effects of Generative AI on hiring - FOMO is not a strategy
Rachel Coldicut gives an assessment of where we are as an industry with making use generative AI at the end of 2024 - AI eats the world
Benedict Evans's keynote from Slush '24 shares his analysis of the state of AI from a business perspective - AI tools for software engineers, but without the hype
Simon Willison shares his relfections in this interview - The expanding dark forest and generative AI
Maggie Appleton's reflections on what is happening/will happen when we flood the web with generative AI content are well worth a watch - “Humans in the loop” must detect the hardest-to-spot errors, at superhuman speed
Cory Doctorow on 'reverse-centaurs' and other AI patterns - Notes on how to use LLMs in your product
Will Larson outlines a mental model for thinking about how to use LLMsin your product - AI for software development: A reality check
Birgitta Boeckeler sharing her explorations of using LLMs to support development at LeadDev Berlin '23 - The LLMentalist effect
Baldur Bjarnason on how chat-based Large Language Models replicate the mechanisms of a psychic's con - The Real Danger of LLMs
Hywel Carver on the risks of using LLMs in our development tools - ChatGPT is nothing like a human
Emily M. Blender on the dangers of anthropomorphising LLMs - ChatGPT Is a blurry JPEG of the web
Ted Chiang share his thoughts around LLMs being lossy text-compression algorithms. - AI Copilot code quality
Git Clear's research into the impact of AI on code quality suggests that people are writing code faster but the quality is going down.
Recommended links about generative AI
Recommended books, articles, sites, talks and podcasts about generative AI.