I think we need to get money and billionaires out of politics. We have to stop trying to distinguish between "bad billionaires" and "good billionaires" based on whether they agree or disagree with our policies. So I'm voting for Anthony

BigTech doing layoffs because of AI gets a lot of attention. I think AI is commoditizing software. Which begs the question what is the future of software companies? Some of the biggest Web 2.0 companies (Facebook, YouTube) were built on the premise that creating webpages was hard.

I've used 2.6 billion tokens with Codex in the past week. This is all /goal for programming and operational tasks. In particular my usage is scaling with human effort as opposed to some other dimension; e.g. asking codex to analyze bluesky message

Google photos titles collages of my dogs "Big blockbuster starring a local legend"

I've been using /goal in Codex for almost every operational task (pushes, run evals, etc...). I think my days of manually typing and running commands is coming to an end. I didn't expect this. A lot of the motivation for investing in notebooks was to allow human editing and execution of commands.

"I’m not the only one. Personal sites are coming back. RSS feeds are coming back. Webrings are coming back. People are remembering that a website can be a home or a place instead of a profile." www.terrygodier.com/the-boring-i... Yessss @terrygodier.com!

SVG is so useful for having Codex render diagrams. Google Docs not rendering SVG is really annoying. Just added HTML cells to web.runme.dev for this use case

Being on call is now codex /goal investigate this alert ; search slack/notion/drive for previous incidents. Produce an @runme.dev notebook with the results of your investigation and suggested mitigations

What does being oncall look like in a world where people can use do million line rewrites overnight?

Living in SF I don't think I fully appreciated the level of AI antipathy building among young people.

First oncall shift since July. What a difference Codex makes. Onboarding to a new service is just so much better when you can ask Codex to explain how it all works rather than stitching the breadcrumbs together yourself.

Gives new meaning to testing on the toilet. openai.com/index/work-w...

Anyone have a good workflow for collaborating with Codex on design docs? Markdown works well when doing single player. But for sharing with other humans Google Docs is better. Codex isn't as good at modifying docs. It can't do suggestions which would be nice for seeing the AI changes.

Does AI make monoliths more attractive? As AI drives software development costs down the cost and complexity of operating microservices starts to dominate. Managing a bunch of middle ware for secret management, networking, etc... seems less attractive than just bundling that into your binary

With WebMCP it seems like the agent is part of the browser or presumably an extension. But Chrome extensions can't start native processes. So if your agent wants to use tools to reach outside the browser users need to install a server separately @bandarra.me @sarahedo.bsky.social

If Trump starts reviewing models I think it will just be another avenue for graft. He's proven with the justice dept and fed he has no interest in independent/apolitical agencies www.nytimes.com/2026/05/04/t...

With AI, the idea that OSS gives you the freedom to make changes is becoming a practical reality. More so for apps than services. #slack comes to mind. The way it gets used is such a dumpster fire the ability to innovate would be a huge boon to the industry.

Initial thoughts on WebMCP. Can the imperative API just be execute this JS code in a JS repl inside the app? Have a mechanism like Pythons help() for the agent to progressively discover and understand the apps SDK. @bandarra.me

I don't think AI will fully eliminate SWEs because great software requires user empathy (hat tip @kelseyhightower.com ). Empathy is the result of lived experience. When you shit the bed doing a demo or taking down prod it sticks with you and influences what and how you build

Maybe the new benchmark for Claude Code and Codex should be how quickly and how cheaply they can clone those apps given access to those apps

Started using web.runme.dev to replace my retool dashboards. Codex does an amazing job of generating static HTML for rendering complex data (OpenAI FineTuning Jobs). It even handled paginating through results

@hamel.bsky.social @sourishkrout.com forked codex is running in the browser. So we can do agentic search of the runme docs. I think I have an AI quality problem. So next step is probably building out some simple evals to optimize the prompts, the docs, and maybe utility libraries

I think safety bound is joining CPU and io bound. With AI we are well into a realm where an AI app could probably do a thing but with non trivial risk of a bad outcome (e.g. data exfiltration)

@tomayac.com what is web-ai-community? Is that a group?

Anyone else trying to get agents working in the browser? I've been working on getting Codex harness to run in the browser. Motivation is to be able to create an AI native UX in web apps.

There's also GrokAI spewing out CSAM.

There was a list of bad AI writing patterns going around the other day; I can't find it now. Anyone recall it?

The war in Iran was started by a democratically elected president. Not a rogue AI.

ChatGPT was pretty unequivocal that threatening to bomb civilian infrastructure can be a war crime. I wonder if you asked OpenAI leadership whether you'd get the same answer? Journalists if your interviewing OpenAI leaders about their industrial policy please ask. chatgpt.com/share/69d51b...

Google Docs/Sheets/Notion/Word/Power Point. These are still the dominant artifacts produced when using AI to answer business questions. They aren't AI native formats. In a post Claude/Codex world documents have two distinct audiences. 1. Humans 2. Ais

Code mode seems to be working in runmeweb. This should fix a bunch of bugs with the AI incorrectly manipulating the notebook. It should also make it easier to extend the AIs abilities just building out the JS libraries rather than needing to add additional tool calls. @sourishkrout.com

If we want to safely sandbox AI actions do we need a low level interface like WASI which we can use to enforce policy and then build rich libraries on top of that interface?

@hamel.bsky.social @sourishkrout.com I think most OSS notebooks (jupyter , marimo) have asymmetric experiences for creators and readers of the notebook. Pre codex/Claude this was fine because most readers didn't have the know how to read/modify the underlying SQL/pandas code.

What is the sharing/publishing story with nteract? If you use nteract to create a report do readers install nteract? Do you use binder or commuter? nteract.io/blog/nteract...

references to Claude code as an operating system; codex plugins; I hope we're not about to usher in a new duopoly of rent seeking platforms. If your not one of the big three I think the move should be to try to commoditize AI.

If your using code/Claude to answer business questions (e.g. how well are you doing selling widgets to millennials) are you producing shareable artifacts answering that question and if so what is the sharing story? Google/notion/word doc? PDf/HTML uploaded to slack?

Is anyone working on using any of V8 isolates, WASM/WASI to create sandboxes for agents? There are two problems I'm interested in 1. Running agent harnesses in the browser 2. Supporting Google Drive as a filesystem Cloudflare blogged about using V8 isolates; anyone adapting that to the browser?

Who's got a good recipe for agentically coding web apps? Specifically producing videos to verify features are working as expected? I've been using @vercel.com 's agent browser. Screencasts vary wildly when running in headed vs headless. Is it better to control recording explicitly?

Does anyone know if Google OAuth requires a client_secret for the PKCE flow? If I don't include a client_secret I get a token exchange fails with 400 client_secret is missing. Is there some user magic words to trigger a PKCE flow without requiring a client secret? My client is for web.

For me codex (app + 5.3) is an inflection point. I have enough confidence in the AIs coding ability to focus more on design and testing. Concretely; iterating with the AI on design docs and CUJs. CI generates videos of CUJ walkthroughs making it easy to verify tests do what they claim.

AI could be the tower of babel for tech companies. Before AI you needed to pool human capital to build something noteworthy. This was a strong incentive to pay the high cost of consensus building. With AI you can get much further by oneself. But that only postponed the reckoning.

As I watch what's happening in Minneapolis and around the country ; to me the immediate risk of AI is that it's consolidating power and wealth in the hands of tech executives that willingly endorse what's happening as long as they get their GPUs.

To all my new England friends. It's 61 degrees and sunny in the bay.

Twitter died a long time ago and X isn't worth it. Before I made the jump, my ego was holding me back. I was too concerned with vanity metrics. I finally realized if you're posting stuff people want to follow, then they'll follow you, even when you leave. www.siliconrepublic.com/enterprise/k...

Police Repeatedly Shoot Tim Cook After Mistaking iPhone For Gun https://theonion.com/police-repeatedly-shoot-tim-cook-after-mistaking-iphone-1824184361/

One of the best things I’ve read in a while:

I work in tech. I emailed my leadership letting them know I hope they acknowledge the events in Minneapolis and take a position. Small as it is; there was a time when I wouldn't have had to ask. I don't know how but I want to try to hold tech accountable.

This is ok tho

I think the market keeps wanting to believe things haven't changed but they have. We are very clearly in an era where the president is willing to go after anyone, any country or any company that displeases him. The volatility should scare most investors. www.nytimes.com/2026/01/21/b...