If you’ve ever opened Chrome DevTools, or optimized a page for Core Web Vitals, you’ve used software built by @addyosmani.bsky.social. We start from when he built a web browser from scratch, at 16, through his 14 years at Google, to what he sees working (and not working) building software with AI:
Addy Osmani
@addyosmani.bsky.social
Former Engineering lead working on Gemini and Google Cloud AI I want you to win :) For more on what I'm up to, I'm at addyosmani.com
"Engineers should own the outer loop. Build agents you can answer for" - My closing keynote for AI Engineer World's Fair is now up and free to check out. www.youtube.com/watch?v=n97B...
Don't Build Agents You Can't Answer For — Addy Osmani
YouTube video by AI Engineer
youtube.com
"The interesting consequence is that the hard part of engineering moved from writing code to deciding whether to trust it, which makes review the most leveraged skill in software right now." More in today's #Radar article by @addyosmani.bsky.social: bit.ly/4oTO571
Agentic Code Review
The following article originally appeared on Addy Osmani’s blog site and is being republished here with the author’s permission.Coding agents are
bit.ly
"Loop engineering is replacing yourself as the person who prompts the agent...However, it’s still early; I’m skeptical, and you absolutely have to be careful about token costs." Check out new #Radar article by @addyosmani.bsky.social to learn more: bit.ly/4oFwiAe
Loop Engineering
The following article originally appeared on Addy Osmani’s blog and is being reposted here with the author’s permission.Loop engineering is replacing
bit.ly
"Long-running agents are what comes next... This post is my attempt to lay out what’s changed, who’s pushing on it, and how an engineer can use long-running agents today without writing the whole thing from scratch." Check out #Radar article by @addyosmani.bsky.social to learn more: bit.ly/4oo0UGA
Long-Running Agents
The following article originally appeared on Addy Osmani’s blog and is being reposted here with the author’s permission.A long-running AI agent can keep
bit.ly
"We’ve spent the last two years arguing about models. Which one is smartest, which one writes the cleanest React, which one hallucinates less. That conversation is fine as far as it goes, but it’s missing the other half of the system." New #Radar article by @addyosmani.bsky.social: bit.ly/43fulk0
Agent Harness Engineering
A coding agent is the model plus everything you build around it. Harness engineering treats that scaffolding as a real artifact, and it tightens every time the agent slips.
bit.ly
A great article by @addyosmani.bsky.social addyosmani.com/blog/cogniti... Offload; don't surrender. "Cognitive offloading is delegating to the AI and still owning the answer. Cognitive surrender is when the AI’s output quietly becomes your output and there is nothing you feel is left to check."
Cognitive Surrender
Cognitive offloading is delegating to the AI and still owning the answer. Cognitive surrender is when the AI's output quietly becomes your output and there i...
addyosmani.com
Caught the great @addyosmani.bsky.social this morning. He along a few others, was featured on a story about Vibe Coding by @bloomberg.com and he had sage statements of course. This man is a legend. You can watch the whole piece : 🎥 youtu.be/xYU7zaaRjmE?...
Cat Wu joined @addyosmani.bsky.social at AI Codecon for a fireside chat on the future of agentic coding and agentic code review, how Anthropic actually uses the tools they’re building, and what skills matter now for developers. Key takeaways on #Radar by @timoreilly.bsky.social: bit.ly/4uqWwID
Everyone's an Engineer Now
Takeaways from Cat Wu’s fireside chat with Addy Osmani
bit.ly
A worthy article on long running agents by @addyosmani.bsky.social - Elucidates the architecture and issues involved: addyosmani.com/blog/long-ru... #longhorizontask #ai #agenticai #agent #agentharness
Long-running Agents
A long-running agent can keep making progress over hours, days, or weeks. It can do this across many context windows and sandboxes, recover from failure, lea...
addyosmani.com
Well-written and immediately effective - and he's got guides to put it into Cursor, Claude Code etc. github.com/addyosmani/a... #ai #agents #coding #cursorai #claudecode Thank you @addyosmani.bsky.social!
GitHub - addyosmani/agent-skills: Production-grade engineering skills for AI coding agents.
Production-grade engineering skills for AI coding agents. - addyosmani/agent-skills
github.com
"The real problem wasn’t messy code. It was that no one on the team could explain why design decisions had been made or how different parts of the system were supposed to work together. The theory of the system had evaporated." New #Radar article by @addyosmani.bsky.social: bit.ly/41tTNSh
Comprehension Debt: The Hidden Cost of AI-Generated Code
The following article originally appeared on Addy Osmani's blog site and is being reposted here with the author's permission.Comprehension debt is the
bit.ly
@addyosmani.bsky.social's multi agent experience agrees with my own, going from 2 to 5 parallel agents is a quick way to transform from a judge to a ticket usher.
Your parallel Agent limit
Running multiple agents in parallel is not just a question of throughput. It is a new kind of cognitive labor that requires managing multiple mental models, ...
addyosmani.com
Old loop open files ➡️ edit ➡️ build ➡️ debug New loop specify intent ➡️ delegate ➡️ observe ➡️ review diffs
Death of the IDE? How Agent orchestration may be replacing the editor as the center of developer work. open.substack.com/pub/addyo/p/... ✍️ my latest write-up
Software is evolving fast. My focus hasn’t changed: helping teams improve performance, UX, and conversion AI generates working code faster than ever… but not optimized for performance That’s where we add value: from “it works” to “it’s excellent” Reading @addyosmani.bsky.social’s new book
"The levels of AI adoption" - Orchestration requires a different skillset. I spoke about orchestrating coding agents at O'Reilly AI CodeCon this week - here's the full free write-up + slides on what makes multi-agent coding works: addyosmani.com/blog/code-ag... Illustration by Steve Yeggae
The factory is a really good mental model to understand the direction of software engineering right now. I am pretty excited about it. Good article by @addyosmani.bsky.social addyosmani.com/blog/factory...
The Factory Model: How Coding Agents Changed Software Engineering
Software engineering is not about writing code anymore. It is about building the factory that builds your software.
addyosmani.com
I finally got around to read this, and oh boy. This is beautifully explained 😍. Thank you @addyosmani.bsky.social 🙏 addyo.substack.com/p/how-modern...
How modern browsers work
A web developers guide to browser internals
addyo.substack.com
How Agent orchestration is replacing the editor as the center of developer work - @addyosmani.bsky.social addyo.substack.com/p/death-of-t...
Death of the IDE?
How Agent orchestration is replacing the editor as the center of developer work
addyo.substack.com
Death of the IDE? How Agent orchestration may be replacing the editor as the center of developer work. open.substack.com/pub/addyo/p/... ✍️ my latest write-up
Death of the IDE?
How Agent orchestration is replacing the editor as the center of developer work
open.substack.com
We talk about the speed AI coding tools give us. We don't talk enough about the hidden cost: Comprehension Debt. addyosmani.com/blog/compreh... ✍ I cover my thoughts on this in a new write-up.
Comprehension Debt - the hidden cost of AI generated code.
addyosmani.com
Cohosted by @timoreilly.bsky.social and @addyosmani.bsky.social, AI Codecon on March 26 is focused on exploring what it takes to build excellent software in the age of AI that creates value for all participants. Register to save your free spot: bit.ly/4bDKyEH
AI Codecon - O'Reilly Media
Join leading developers and technical experts as they share how AI is shaping new workflows, tools, and ways of building software.
bit.ly
Announcing Gemini 3.1 Pro! Google's smarter model for your most complex tasks just shipped 🚀 A few web demos:
AI users are working longer because, for the first time, our tools can finally keep pace with our imagination. When output is cheap, your greatest leverage isn't how much you build, but the taste to know what is worth building - and the rest required to see the difference.
AI writes code faster. Your job is still to prove it works. - addyosmani.com/blog/code-re...
AI writes code faster. Your job is still to prove it works.
AI did not kill code review. It made the burden of proof explicit. Ship changes with evidence like manual verification and automated tests, then use review f...
addyosmani.com
Liked this observation from Addy Osmani: “Abstractions don’t remove complexity. They move it to the day you’re on call.” addyosmani.com/blog/21-less...
21 Lessons From 14 Years at Google
Lessons learned from 14 years of engineering at Google, focusing on what truly matters beyond just writing great code.
addyosmani.com
Excited to share my new YouTube series on Software Engineering Effectiveness! www.youtube.com/watch?v=lqcK...
Software engineering: efficiency vs. effectiveness
YouTube video by Google for Developers
youtube.com