Post #529732

2026-01-12 19:26 UTC

LLMs are -- to put it euphemistically -- seeing increased use among software engineers. This has sparked much consternation about what this means for our engineering rigor, but we have found that prudent use of LLMs can, in fact, allow us to be much more rigorous in our engineering. Today, @ahl and I will be joined by Oxide engineers who have been at the forefront of using LLMs as tools to increase the rigor in their own work. Join us, 5p Pacific today! https://discord.gg/QrcKGTTPrF?event=1460344349857546367

Replies (6)

@charlag@birb.site 2026-01-12 19:47

@ahl@mastodon.social 🙄

Open ##2947926
@keyboardg@mastodon.social 2026-01-12 20:15

@bcantrill@mastodon.social @ahl@mastodon.social Not sure I can make the call, but on my mind has been if there is longer term loss in pride of craftsmanship with reaching for LLM solutions. Even slipping care for the user needs in turn. Maybe this feeling is influenced by the noise from llm vendors. So many of my breakthroughs have come from exploratory coding. Lightbulbs that go off for future work.

Open ##2947927
@JohnAZoidberg@mastodon.online 2026-01-12 20:21

@bcantrill@mastodon.social @ahl@mastodon.social Excited to hear your thoughts again! I had only vibe code by chatting with Gemini previously, until I heard the 2026 predictions episode. Now I am trying out Claude and the second thing I had it help build me was a new test suite, which I had wanted to build for 2 years. I thoroughly reviewed the test cases and now either I or AI can more confidently continue making changes.

Open ##2947928
@mjd@mathstodon.xyz 2026-01-12 20:30

@zeenix@toot.cat I was very impressed by this Oxide policy statement, ”Using LLMs at Oxide”, and read it more than once. Maybe you will enjoy it. https://rfd.shared.oxide.computer/rfd/0576

Open ##2947929
@n_dimension@infosec.exchange 2026-01-12 20:44

@bcantrill@mastodon.social @ahl@mastodon.social Using LLM is a learned skill. Great to see engineers leading the way.

Open ##2947933
@mwcz@mastodon.social 2026-01-12 22:08

@bcantrill@mastodon.social @ahl@mastodon.social Looking forward to it! I'm especially curious to hear ways to stay rigorous when reviewing LLM output from others. Whether it's PRs, job applications, documentation... the sheer volume of these things is increasing rapidly and staying rigorous is getting more challenging.

Open ##2947934