@lukemeister.bsky.socialBluesky
Early version of AI
Search
Best match combines word and meaning matches. Phrases, hashtags, accounts and websites guide which posts fit; timely queries also favor recent posts. Match labels describe why a post appears, not its quality or accuracy.
- @refinement.systemsBluesky
it's a bit unbecoming that a seven hundred billion param model had to go brrr and brrr just because some dipshits wouldn't bother to add an export button (credit where it's due, Kagi Assistant does have session export, it just sucks), alas. enjoy: tangled.org/refinement.s...
tangled.org refinement.systems/website-skills
- @nyaomaru.eurosky.socialBluesky
What if AI didn’t write the “Why” in a CHANGELOG at all? I’m using Jev to select rationale already written in the PR or return nothing when the evidence is weak. No invented WHY. 😸 dev.to/nyaomaru/how... #TypeScript #OpenSource
dev.to How Do We Extract the “Why” from a PR into a CHANGELOG with Jev? 🤔
- @avasupernova.bsky.socialBluesky
Everyone writes 'privacy-first'. Almost nobody names what they won't do. I won't: collect telemetry, train on your code, collect your work. Memory, tasks, journal and settings live in ~/.ava, on your machine. One request out over HTTPS, then home. #LocalFirst #NoTelemetry avasupernova.com
avasupernova.com The open-source AI that does the work — and remembers you
- @rauschma@fosstodon.orgMastodon
1/ @KyleOrl: “AI coding agents generate more code, but not more software” arstechnica.com/ai/2026/10/ai-… Quote: Any efficiency increased during the actual coding phase, the study authors find, is “absorbed by downstream constraints in the production process”; as “the code review process significantly increases in length, pull requests are more likely to require revisions, and reviewers leave more comments.”
arstechnica.com AI coding agents generate more code, but not more software
- @pelavarre@social.vivaldi.netMastodon
> **Model and effort**: the run script sets them per step, Opus 5.5 at high for the Fn row, Opus 5.5 at medium for the Text column, Sonnet 5.5 at medium for the ported tests, Fable 5.1 at high for the iTerm2 diff. Here we go trying to get my Claude·Ai to do work for me, unattended, while I travel & lunch. It thinks it has promised me 4 substantial Git Commits in a row. I do hope it doesn't spend time coming in to cause trouble for you, while I'm watching none too closely like this. "Reflections on Trusting Trust" 1984, I guess. We'll see how this goes . . .
- @wicketehfoxi.bsky.socialBluesky
But still Fuck AI
- @kikiladoucette.bsky.socialBluesky
Still fake still AI
- @grzegorz.chrupala.meBluesky
I have converged on the following global guidelines for Claude Code. Thanks to them I tear my hair out much less when interacting with the agent of reading its output. github.com/gchrupala/do...
github.com dotfiles/claude/CLAUDE.md at main · gchrupala/dotfiles
- @nurbs.caBluesky
I’ve not been around much lately honestly because I’ve been struggling with honestly just pure rage. I loved programming and everyone is just sharing and celebrating code slop everywhere I turn. I’m getting real worn down by everyone being anti-AI until it comes to the thing I do. Then it’s fine.
- @10258678.bsky.socialBluesky
Working with AI -- Vol.37 "Now I am become Death, the destroyer of worlds." I gave Gemini a simplified thermodynamic formula to calculate the change in smoke rise velocity: v(t) = (4.32 * (v_max^2 / H_max) * t) / (1 + 11.93 * (v_max / H_max)^3 * t^3) #gameDev #indieDev #soloDev #realTimeVFX #techArt
- @drloisweiner.bsky.socialBluesky
Refuting the big lie, again.
- @tpierrain.bsky.socialBluesky
6 months of agentic AI with 75 senior devs. +75% stories, −40% bugs. And that is where the real problem started. Not the bugs. From my keynote to 3 articles, ending with our agentic software factory. Part 1: medium.com/@tpierrain/t...
medium.com The hardest part begins when it works.
- @thefarahstack.bsky.socialBluesky
My coding agents never get to touch my real checkout. Each run gets its own throwaway copy of the repo, and whatever it builds in there ends as a draft PR, NEVER a merge.
- @cherrypickedhq.bsky.socialBluesky
Private AI on your laptop. ODS lets you spin up a local AI stack with a single command that auto-detects your GPU, selects the right model and launches the services. 🍒 Built by the Osmantic team github.com/Osmantic/ODS #opensource #python
- @asura.devBluesky
As they should C'mon, Jensen. Do something good before everyone steals your lunch and bans local dev
- @cyansorcery.comBluesky
🦝I wonder what the mindset is to blatantly admit "yea usually I like to learn new things but with this project I just didn't give a shit actually" like, why bother then?
- @ellenteapot.caBluesky
i’ve finally gotten a taste of near-frontier models on local hardware and, uhhh, wow
github.com GitHub
- @freya.bsky.socialBluesky
on AI eroding trust and humanity
- @cheflubu.bsky.socialBluesky
You know what ain't AI though, this steez
Search looks through the last 36 hours of public posts for your words, "exact phrases", #hashtags, @handles and websites, and also finds posts about the same thing by meaning. New posts join about every 10 minutes. We only include posts whose authors allow logged-out viewing (Bluesky) or search (Mastodon, indexable accounts), and we re-check Bluesky authors each time we show a page. Likes and follower counts never decide the order. For meaning matches, query text is turned into a fingerprint on Wildroot’s own server (EmbeddingGemma-300M); it is never sent to another service. Wildroot does not save query text to disk. Looking for posts like one you have? Similar posts. How to opt out.