Copilot Platform @GitHub. Formerly Observability @Honeycombio and F# lang/.NET tools @Microsoft

Redmond, WA
Joined August 2015
Work laptop has been properly adorned
2
11
1,781
Okay this one was kinda funny
236
I don’t think personal software will kill OSS and I think it’s very overrated. It’s super fun to jangle the keys in front of your own face and vibe code up a copy of something else that exists in this world. But truly owning it, operating it, having a vision for it kinda sucks! OSS still wins out on the a value proposition that it’s free and you can shunt those concerns off to someone else. The talking computers don’t actually know exactly what individual people want and most people don’t feel like just living with whatever an agent confabulated as a design and architecture. The future for OSS probably involves a token market where you can yield some tokens to allow a maintainer group make the right decisions for you.
4
2
1
23
927
Bo Nix is the John Elway of Tim Tebows and it’s beautiful
Truly disgusting player
1
200
More seriously, Nix has that fourth quarter dog in him. Stats don’t show that, but his winning record does
114
Introducing my new side project - Fai. github.com/Krzysztof-Cieslak… It’s a functional programming language with which I try to explore how we can design better “classical” software engineering tools specifically for AI agents. A couple of design choices I've made are described ⬇️
13
21
4
210
24,391
Daily driver is now Argon, it’s great
1
3
460
Phillip Carter retweeted
What should we improve in the GitHub Copilot app?
133
13
87
21,688
So far Argon really does seem the best at absorbing a bunch of text and turning into a rough draft of a doc with decent writing style. But man, so much steering still. AI assisted doc writing is like working with the first coding agents when they came out
2
7
804
Like you tell it no fluff, grammar of a fifth grader, ELI5, simplify, keep it terse, brevity, all the intent-based keywords… …and every one of these damn tensor swamp pits just jumps at the bit to slop you up some extra wordy bullshit
1
2
197
Cursor’s de-slop doesn’t work, none of the skills I try work, I dump all my own writing in as examples and it doesn’t work, it’s still just so mid! But Argon definitely feels like it made progress over the rest I think
1
181
Phillip Carter retweeted
New on our Engineering blog: Creating in-house source control. In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale. Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: anthropic.com/engineering/re…
1,173
2,658
2,177
25,920
8,215,279
Has ChatGPT on iOS gotten super slow for people? iPhone 15 not on low power mode so I don't think it's my phone. Takes ~3-5s to load in the sidebar the convo I started on desktop and got a notification for on the phone. Then noticeable lag as the keyboard popped up.
1
679
bringing back this classic
i am so stressed out from running my pet lizards instagram account i’m about to have a nervous breakdown
1
4
493
It's really quite something when doing stuff with LLMs that you have to oscillate between deliberately built verification systems and just yelling at your little demon servant differently
2
7
480
Yeah I think this is pretty much the most compelling argument. Dev managers have to fight to create budget and space for people to experiment with AI if they care about their peoples' long-term careers.
🗞️ if your team is too busy doing their 'normal job' to experiment with AI, you're preparing them to be replaced ghuntley.com/replaced/ In this post, I guess this's the last time I'll say it. AI use is no longer optional if you wish to be employed. What follows is advice for current employees and hiring managers on how to hire in 2026 to source AI-first candidates.
1
312
So I think this might be a smart way to just get unstable people off your platform as a way to decrease liability risk
Anthropic may start banning people for bullying Claude Their upcoming Usage Policy prohibits “sustained and needless abusive or cruel behavior toward our models”
1
2
344
Basically the goal isn’t to be productive. That’s for lame stuff like work. Instead it’s a fun game to try and speak the magic commands to the tiny tensor demon on your laptop and keep trying until you find one that works
106