2 HN users recommended this demo2 months2025–2026score 2
Years with an HN recommendation, 2007–2026
Pointed to as a non-promotional, realistic long session of an experienced engineer working with AI on non-trivial code.
Jon Gjengset spends about three hours livecoding complex changes in a Rust codebase with agentic AI tools, testing his hypothesis that AI is weak at non-pattern work. He documents where Claude Code helps and where it struggles.
“I found this one and sat down and watched the whole three hours.” (HN, 2026)
Jon Gjengset (jonhoo) Who is famously fastidious did a stream on live coding where he did something similar in terms of control. Worth of a watch if that is a style you want to explore. [this video] I don't have the energy to do that for most things I am writing these days which are small PoC where the vibe is fine. I suspect as you do more, you will create dev guides and testing guides that can encapsulate more of that direction so you won't need to micromanage it. If you used Gemini CLI, you picked the coding agent with the worst output. So if you got something that worked to your liking, you should try Claude.
Ironically, when machine learning is getting “deeper & deeper”, human learning is getting more and more impatient and shallow. I have been searching “Vibe coding” videos on YouTube that are not promoting something. And I found this one and sat down and watched the whole three hours. It does take a lot of real effort. [this video]
Non-promotional vibe coding session worth watching in full
This video [0] is relevant, though it actually supports your point - it shows Claude Code struggling with non-trivial tasks and needing significant hand-holding. I suspect videos meeting your criteria are rare because most AI coding demos either cherry-pick simple problems or skip the messy reality of maintaining real codebases. [0] [this video]
Shows Claude Code struggling with non-trivial tasks
take a look at the replies to [link] eg [link] and [this video] PS, what was your last setup? performance depends a lot on the domain/model/workflow[1]; some combinations don't work at all [1] by workflow I mean: trying to oneshot a feature VS interactive exploration VS boilerplate generation
Example video of developing non-trivial things with AI
Video list and Hacker News comment labels from HackerTube (prompted by @karpathy), released under ODC-By 1.0, built on the open-index/hacker-news archive. Hacker News data is provided by Y Combinator; comments remain the work of their authors. Video titles, channel names and thumbnails belong to their creators and are shown via YouTube. HackerTube Watch is an unofficial edition. Data export of 2026-10-31.
Plays via YouTube's privacy-enhanced embed
4 Hacker News comments
From the HN archive; up to 20 kept per video, favoring different commenters. Recommendations first.
Read more
Citations, criticism and jokes