Will retweeted
I’ve been working on something new: aiforui.dev/
AI for UI has 28 skills and 20 lessons on building great interfaces with AI.
More than 3,000 people have enrolled so far. It’s $299 instead of $399 until early access closes on Friday.
The full course opens in 2027.
Will retweeted
The MotionSites skill is live 🚀
I turned 700+ of my best website designs into one skill.
Install it and your agent stops making generic AI slop and starts building cinematic, animated sites like these.
Works with Claude Code, Codex, Cursor or any agent.
Free and open source:
github.com/Motionsites-AI/cl…
Will retweeted
Finally summarized what I've been sending to friends in DMs View article
Will retweeted
> Don't wait for your agents. Let them wait on you. If you worry about TTFT, you're probably doing something wrong. Lean into async. Give them bigger tasks, be more ambitous.
Yes
Finally summarized what I've been sending to friends in DMs View article
I'll be speaking at this event organized by @exedev and @jana_iris at @sentry next week to share a bit about what we've found effective when working with agents and maybe some idiosyncratic viewpoints and demos. Hope to see you there!
luma.com/0qr6bxcw?tk=PM7pb4
Will retweeted
No pull requests, no code review. We push to main.
Our agents read production logs, watch deploys, grant feature flags, and check their own work end to end.
Here's how we do that:
Will retweeted
This came from a big customer onsite last week and it went over really well. Of course, Thorsten uses Amp as the vehicle, but the talk is really about the meta of using agents today, what patterns are starting to emerge, and where things are headed
GPT-6 and Intelligent UI, now rolling out in ChatGPT for everyone.
Intelligent UI in ChatGPT delivers fast, interactive answers that make everyday questions more visual, complex topics easier to grasp, and interactive tools for your task available on the spot.
You don't need a PhD to run evals.
@sqs had agents build fake shopfronts and permit sites, complete with annoying cookie banners, to improve browser use in orbs.
“It's like with usability testing: you learn most of the insights from the first three people you talk to”
Would you prefer to save money with dial-up when your competitors are using fiber?
Recorded on Hot Take Thursday, @sqs and @thorstenball are back with a new Raising An Agent. In this episode they discuss how cheaper models end up costing more, why they do not not wait for agents, running evals in Orbs, and the hot takes Thorsten left out of his X post last week.
00:00 Intro: Engineers who forget the history of technology
01:55 Hot Take Thursday & a €50-a-week token budget
05:14 Why cheap models cost more than frontier models
08:11 Token budgets & who gets priced out
14:06 Disrupt yourself with small teams & unlimited tokens
16:38 Agents go horizontal across the software lifecycle
20:56 Rollouts, bug reports & "I'm feeling Pucky"
23:53 Code review, monorepos & other walls around the agent
25:30 How to show a cautious company what agents can do
29:48 Why most people haven't seen what agents can do
31:57 Don't wait for your agent
33:38 Simple evals & agents testing agents in Orbs
38:27 Why Amp has runners too
40:47 Follow your energy & stay aligned
46:33 The hot take Thorsten left out
54:14 Do languages & frameworks still matter?
Will retweeted
We gave Claude Opus 5.5 access to 12 laptops.
The task: turn one prompt into a complete launch video.
Using Computer Use and Higgsfield MCP, it:
> divided the work across every computer
> created the visuals, built the animations
> assembled a fully editable After Effects project.
This is what an AI-native motion studio looks like.
Will retweeted
Introducing Eleven v4 and Eleven v4 Turbo, our fastest and most emotive voice models yet.
Ranked #1 by Artificial Analysis.
This has never been more true. The opportunity cost of endless planning and contemplation just went way up. Agents allow you to just try vastly more stuff, so let them, and you'll learn way more, way faster.
This quoted post is unavailable.
Will retweeted
My god this is such a good speech that every SWE needs to hear. You know what? Every person should hear it
Keep the happy memories, eyes on the reality, be excited about the future. That’s the best that anyone can do
"Just use a cheaper model" sounds like saving money.
But fiber didn't just make your internet faster. It made whole new businesses possible. @thorstenball and @sqs discuss how the same is true with frontier models.
Would you prefer to save money with dial-up when your competitors are using fiber?
Recorded on Hot Take Thursday, @sqs and @thorstenball are back with a new Raising An Agent. In this episode they discuss how cheaper models end up costing more, why they do not not wait for agents, running evals in Orbs, and the hot takes Thorsten left out of his X post last week.
00:00 Intro: Engineers who forget the history of technology
01:55 Hot Take Thursday & a €50-a-week token budget
05:14 Why cheap models cost more than frontier models
08:11 Token budgets & who gets priced out
14:06 Disrupt yourself with small teams & unlimited tokens
16:38 Agents go horizontal across the software lifecycle
20:56 Rollouts, bug reports & "I'm feeling Pucky"
23:53 Code review, monorepos & other walls around the agent
25:30 How to show a cautious company what agents can do
29:48 Why most people haven't seen what agents can do
31:57 Don't wait for your agent
33:38 Simple evals & agents testing agents in Orbs
38:27 Why Amp has runners too
40:47 Follow your energy & stay aligned
46:33 The hot take Thorsten left out
54:14 Do languages & frameworks still matter?
Will retweeted
sharing some notes on typesafe 🤝 coding agents:
docs.google.com/document/d/1…
we likely will never have time (ever again) to play ourselves, but hope the that the community goes WILD (and makes me look like a naive idiot)
Will retweeted
This blew up beyond anything I expected.
The responses are surprising too:
40% says that, duh, this is already the reality at my job. Yes, at mine too.
Another 40% agrees with some adjustments and different timelines.
Then 20% disagree. Of those that disagree, some, of course, call me an idiot and that I only build CRUD apps and so on. Not surprising, it's the Internet after all.
But there are two other categories of criticism that I find fascinating.
One is that people say I work on AI and that is why I would predict AI to take over. The shovel seller says you should dig, is the criticism. I get it. But in my case they have the causality backwards: I work on agents, and have done so for nearly two years now, because I thought and still think that this is the biggest lever for developer productivity I've seen in my career. I'm talking about shovels and am in the shovel business because I think everyone should have one, not the other way around.
The other critique is... strange, somewhat: people get offended by the predictions I shared here and get mad at me as if these were not predictions but wishes.
Listen: I do not wish for anyone to be priced out of building software when I predict that some people will be. The weatherman doesn't want it to rain when he says there's gonna be showers on Tuesday.
I'm merely predicting what I think will happen.
I didn't say it's good or bad. I didn't say PMs shouldn't exist or Italian shoemakers.
I'm making a prediction on what I think is going to happen, that's different from wishing what will happen.
That being said: I wasn't always delighted about how things were going. I loved typing out code (I'm faster than you, pal), I loved debugging by hand, I loved my editor and my shell and my color themes. And AI and agents took a lot of that away from me.
But I also hate being inefficient and 1.5yrs ago it became pretty clear to me that to continue writing code by hand will be inefficient and a waste of time. So I adapted.
And now, with the latest generation of models, I'm having a ton of fun again: Fabel and Astra let me do things I never would've done by hand and now they also do the shitty parts and don't leave me to do the remaining shitty 10% to get it shipped.
I'm excited and optimistic and think there's a ton of stuff to build and now we have superpowers that let us build more, faster; superpowers that truly let us shoot for the moon.
Jev + GrokBot is the best AI agent system I’ve built in my life
It just made my setup CHEAPER and FASTER than what 95% of people are running...
setup takes literally 7 minutes:
prompt → GrokBot → Jev decision → GrokBot execution → result
step 1 → open @typesafeai , create API key (keep it off chat paste)
step 2 → tell Grok Bot: store TYPESAFE_API_KEY in the secure field
step 3 → prompt Grok Bot: install typesafe-sdk on Agent Computer + smoke system_one (Choice)
step 4 → tell Grok Bot: build the usage lab (router, dry-run, config, logs) - or clone Github below
step 5 → add skill jev-usage-router: before browser / research / retry / extra bot → call the router, honor action
step 6 → stay shadow first, read logs, then active when you trust it - kill switch: bypass jev or enabled: false
step 7 → flip active: GrokBot obeys route - Jev decides - GrokBot executes - humans control irreversible actions
the result: Jev + GrokBot the best and fastest agent running directly on your computer rn, I’ve already tested it on routine tasks - and the results are genuinely incredible
You can come up with endless ways to use Jev + GrokBot - but the most important thing is to install it as soon as possible
Copy this 2028 setup, explore my repo below - then read the full Jev deep dive ↓























