Ellie Huxtable
@ellie.wtf
2 months ago It’s been so so nice! 😊
💬 0
♻️ 0
❤️ 1
It’s been so so nice! 😊
Travelling while gay: “We upgraded your room for you and your friend” 😭
AI use case: Code optimization
AI can automatically improve your code for speed, efficiency, and readability. It helps you ship faster, reduce technical debt, and make your projects stronger from day one.
Which part of your code would you optimize with AI first? i love that two people have already made visualizations of this on @observablehq.com observablehq.com/@roop-pal/gr... observablehq.com/@jwolondon/g...
i think there's still room for making a spinny interactive demo but on the other hand i should maintain a sleep schedule gray codes in rotary encoders look like ancient geometric puzzles from sci-fi, translating an angle of rotation to a number in binary en.wikipedia.org/wiki/Gray_code not yet but slowly writing a post about it in my head
@chrisdowney.bsky.social can you please send me a DM, i'd love to chat, it's about GitHub. Nahrin Jalal is running a workshop at Cursor Colombo’s International Women's Day virtual co build on March 8.
Come build with her, then deploy live with Netlify (fast deploys + Preview Deployments). Free credits.
luma.com/iwd-buildwcu... 👀 Built something cool with Prisma?
🙌 Share it in the community-showcase channel on our Discord.
It’s where developers post real projects, give feedback, and learn from each other’s work.
Join the server and drop your project 👇
pris.ly/discord/var-3 I am going to start posting more here. What kind of database content do people want?
Vibe coding is growing up so fast. 💨
Tomorrow (12pm ET), Karthik Puvvada (Netlify) sits down with Eric Simons (Bolt.new) to talk AI agents, the new builder workflow, and why AX-first is the difference between “cool demo” and “shippable product.”
⌚️ Set the reminder: www.youtube.com/watch?v=RNfu... Taste is the biggest engineering moat for you & your team.
Alhamdulillah!
Let's go!! I built my first coding agent called `clai` in July 2020 early GPT-3 access via Greg Brockman. 3yrs before ChatGPT. A year before Copilot.
This is what I've been building toward.
Command Code is now live!
Sign up and get $10 in free credits.
↳ CommandCode.ai I'm Ahmad. 25 years writing code. Contributed code that flew on Mars with NASA. Built 300+ open source cli pkgs. Got Meta to open-source React. Google Developers Advisory Board (gDAB) founding member.
↳ github.com/AhmadAwais
↳ AhmadAwais.com/about Chowdhary, Evil Rabbit, Ahmad Nassri, Taimur Rashid, Sanjeev Sisodiya, @farazamiruddin, Britton Winterrose, Aravind Putrevu. Price, David Thyresson, PWV (Preston-Werner Ventures), Firestreak Ventures: Walter Kortschak Amir Rustamzadeh, Mento vc: Alex Zhuravlev Dmitry Golomidov, Sam Awrabi, Banyan Ventures, Igor Ryabenkiy AltaIR Capital, ENEA: Amirteymour Moazami, Alumni Ventures, Paige Bailey, Ian Livingstone, Anand Guy Podjarny (CEO, Tessl, Founder, Snyk)
Theo Browne (CEO, T3Chat, YouTuber)
Logan Kilpatrick(Google/ex-OpenAI)
David Mytton (CEO, Arcjet)
And many more on our about page
↳ commandcode.ai/about
PWV: David S. Our $5M seed is led by PWV fund of Tom Preston-Werner (Co-founder of GitHub).
Joined by incredible investors and founders:
Luca (CFO, Apple)
Dane (CTO, Cloudflare)
Paul @kiwicopple (CEO, Supabase)
Amjad Masad (CEO, Replit) & Michele Catasta
Feross (CEO, Socket)
Zeno Rocha (CEO, Resend) Your taste is transparent and stored in a readable taste[.]md file.
You can inspect it, edit it, or reset it entirely.
And you can share it:
$ npx taste push --all
$ npx taste pull ahmadawais/cli
Senior engineers can encode their patterns for everybody on the team. The benchmarks speak for themselves:
CLI scaffolding: 4.2 edits → 0.4 edits (month 1)
API endpoints: 3.1 → 0.3
React components: 3.8 → 0.5
Test files: 2.9 → 0.2
The improvement compounds.
More usage = better constraints = fewer corrections. "But what about rules files?"
Rules decay. You wrote them six months ago. Your codebase evolved 100 times since. Now they contradict each other.
Command learns continuously from what you actually do, not what you remembered to document.
Taste is auto-learned/managed/updated! How? A new AI model architecture `taste-1`.
It's a meta neuro-symbolic model that separates learned constraints (symbolic) from generation (neural).
Your interactions become reward signals. The symbolic layer updates in real-time. No fine-tuning. No batch training.