Dan Barclay (@DanielBarclay) profile picture

Software Engineering Manager 🧑🏻‍💻 / Ending food waste for good https://xitter.gq/t.co/gm4yqFjYJb / The CMS built for AI https://xitter.gq/t.co/hTAhlVMCzd

London, UK
Joined May 2011
Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)Photo by Dan Barclay (@DanielBarclay)
Loving building on @Cloudflare as my go to platform for deploying pretty much everything. Seeing them lead the way with MCP, the collab with Astro, TanStack to power open source and the ease of use vs AWS makes it a no brainer for me in 2025
9
4
82
10,900
Being able to interact natively with your app in the simulator/xcode might be the only acceptable reason for devs to get a touchscreen MacBook
Sorry, but if I see you doing this publicly I might smack you.
1
34
Asking Claude to create a pr-and-commit-style skill, made on how I used to handwrite PR descriptions and commit messages (before AI) has been a total game changer in the fight against slop github.com/barclayd/barclayd…
1
98
MacOS on the iPad/iPhone Duo when?
Here are other optimizations for touch on the new MacBook Pro Swipe up for dock like on an iPad New animation for drag & hold in Finder
46
This is diabolical 😂
Native iPhone Duo apps, built with Expo. Ship yours before your AppleCare appointment!
18
Codex looks outstandingly good for DDD (Duo driven development). Can’t wait to get my hands on this
I can confirm codexmaxxing on the Duo is a thing!
1
3
619
Vibecoding my way to an iPhone Duo after @EE promised a launch day delivery and then retracted it the day before launch
4
2
607
Dan Barclay retweeted
Here's the new Claude experience on the iPhone Duo! It's going to be amazing for multitasking, Claude Code, and creating new things. Looking forward to see folk start using it tomorrow!
274
290
153
7,199
1,027,880
GPT-6 Astra is also exceptionally good at making app icons with Computer Use 🔥
87
Just did the same for the first time in the 8 years I've had a GitHub account. Truly is a sign of the times
Replying to @kentcdodds
My biggest friction now is the volume of GitHub emails I'm getting. I may have to turn off email notifications from GitHub 😳
1
124
‘You can just do things’ is the way every dev should embrace building software in 2026
307
Try promptlycms.com for free today!
Replying to @mattpocockuk
@mattpocockuk I’m an indie dev who got fed up of having prompts in source code, so I made a solution that integrates with AI sdk to make it easier for teams to manage. It’s free and open source. I’m a big fan of your work and would love your feedback pls! promptlycms.com
2
80
Dan Barclay retweeted
i still think seoul station has one of the goated UX
58
780
171
7,768
1,218,076
Dan Barclay retweeted
It's pencils down, people. Writing code by hand is no longer an economically viable skill for most programmers at most companies. But the future of making software has never been brighter. Don't you dare black pill this beautiful moment! youtu.be/vDjW_dRyKXY?si=6Fsf…
417
990
517
9,241
3,187,143
Deterministic validators > harnesses for AI
Introducing shadcn/lint. An agent-first linter for Tailwind design systems. You define what’s allowed. When an agent breaks a rule, the error explains what’s wrong and how to fix it using your components, variants and theme. There’s a lot you can do with this. Let me show you ↓ github.com/shadcn-ui/lint
2
58
❤️
24
3,673
64
35,209
819,264
I've had the exact same experience. The change of behaviour and inconsistencies have resulted in me changing to Codex
I’ve been using Claude Code since July of 2025, and have been on the Max plan the entire time. There was a solid 6 months when Claude was fantastic. The models worked great. And things could be predicable. After 4.6 things started to change, and known behaviors started changing. That’s not bad, models change. The importance here is the model didn’t ~regress~ on my own workflows and integrated seamlessly-enough that it didn’t bother me. 4.7 has changed that. Ever since its release I have fought this model. On behavior. On knowledge. On ~laziness~ especially, something the prior models I found for my tasks had no such issues. This model is the first I cannot rely on. Yes, insert one should never rely on a 3rd party model, but for those of you working this industry you know we still do to some extent. This will be the only tweet I do this rant, but after 10 months of using Claude, I’m gone.
1
115
Dan Barclay retweeted
me and Claude after we took down prod together
me and him after the dumbest fight but we can’t stay mad at each other
44
1,165
15
19,473
835,366