Alex Sorokoletov

AI coding tools are perhaps our new terminal emulators So I'm currently over in San Francisco. I've been here for almost two weeks now. I'll be heading home to my family in a couple of days. But over the weekend, I had the opportunity to drop into the Com...

ghuntley.com/vt100/

LLMs are not ready for this: HTTP 500 OK

the six-month recap: closing talk on AI at Web Directions, Melbourne, June 2025 Welcome back to our final session at WebDirections. We're definitely on the glide path—though I'm not sure if we're smoothly landing, about to hit turbulence, or perhaps facing a go-around. We'll see ...

ghuntley.com/six-month-re...

Now there is so much stuff to do, what to pick ...

I finally got my first AI-native dev loop working yesterday. Based on the assumption that I accept all results agent is doing and I can only control inputs (prompts essentially). Head is spinning from the fact that this is possible _now_ and we are so early into this.

Great recap, shared with my circle. Feeling the same trend and seeing the thoughts structured nicely by someone else, and compare with 6 mo ago is both a proof and concern (am I 6 mo behind - ages in AI world).

Yep, he seems to be non-stop busy, working on good content! We all can wait, I guess :) Thanks for being busy too, @ghuntley.com!

ok this one was enjoyable too. not sure i have the patience to dig into exercises but maybe i should. pytorch looks usable and not too scary

in the grander scheme of things im not sure how useful it is to dig into these fundamental. i guess im partially curious and partially fomo

Even though I get a gist and short summary, it still seems helpful somewhat in building agents

Agree with the patience aspect. Maybe rephrasing it with "enough time" even. Started using NotebookLM for the interesting papers, videos and articles that I would otherwise completely skip. Instead I can walk and listen 10m of summary.

the z80 technique reveals the source code for Atlassian's 'rovo' AI assistant Ever wondered what happens if you take the technique at "Can a LLM convert C, to ASM to specs and then to a working Z/80 Speccy tape? Yes." and run it against the Atasslian Command Line (ACLI) interfa...

Impressive work by @ghuntley.com
ghuntley.com/atlassian-ro...

When is he coming to @gergely.pragmaticengineer.com podcast?

Just as surprised as anyone that Mashable or MSN seem to no longer be reliable: they publish fake stuff quoting influencers making things up, and never bother to correct.

Like this Builder .ai story about “devs pretending to be AI”

Their incentive is views/ads not correctness

TypeScript and Python SDKs for Claude Code are here.

Use the same agent and tools as Claude Code, and build them into your own agents and workflows.

YouTube video by Anthropic A conversation on Claude Code

If you’re new to Claude Code, here’s Alex Albert and I discussing its origin story:

www.youtube.com/watch?v=Yf_1...