erik

alright all of you that maintain a cli oauth flow i hope it's obvious to you now doing the whole browser link callback to localhost thing is dumb and annoying af in ssh please implement the code flow that polls - try gh cli login flow to see it

100% this. Little chunks of reusable content you can use to steer your friendly clanker make a big difference.

Your AWS bill once you put a tokenmaxxing leaderboard up for the engineering team:

@dusterbloom @malikwas1f @davideciffa Between capability nerfs and usage limits local agents can’t happen soon enough!

@malikwas1f @davideciffa @easel Mate, appreciate and actually been working on it but Opus is so nerfed it is unbelievable ... can't get it to finish HumanEval+ ... fricking incredible

Huge thanks to @dusterbloom and @easel for implementing prefix caching + cold-start tuning and the upstream CUDA VMM fix. Luce PFlash is now ~10× faster warm, ~2.5× faster cold (block sparse attention autotune). Live for Qwen3.6 27B! 🏎️ github.com/Luce-Org/luceb…

Every time we engage with an influencer on X dot com, Dell sends them an XPS. Anyone want a free Dell XPS?

@pupposandro Got it running on my RTX 5090 Laptop under WSL. Awesome! Next step, getting it wired up to ddx agent.

Qwen3.6-27B at 35 tok/s on a GB10 DGX. Almost 3× faster than vLLM+DFlash, 9× vs vLLM bf16. Luce DFlash is now available on Blackwell consumer GPUs. 5090 and GB10 owners, you've been asking. OpenAI-compatible tool calling works out of the box, so it drops straight into OpenCode, Hermes, Cline, whatever you run. Huge thanks to the incredible @superoo7 for shipping this to the community. Repo in the first comment.

@Steve_Yegge Teams limits are way below Max 20x and not pooled very well at all, from my anecdotal experience. Off hand, I'd guess that the $150 teams seats might get a similar budget to the $100/mo max plan.

This is my third $200/month Claude Pro Max plan. My first two have maxed out their weekly limits, and now I'm about to hit a session limit in my third one. And I'm trying to dial UP my usage, with Gas Town. Has anyone done the math to figure out when it becomes more economically sensible to buy the Teams package and give yourself all five seats?

I've been using gastown. This may not be the final iteration, but it is the way.

Managers have been vibe coding forever by @thekitze :D

On the ground at SFO. I’ll be here all week attending #inbound2025. If you want to talk email and AI hit me up!

@system76 Slackware, 93 or early 94. 386-40 and 386sx. Booting Linux into 2MB RAM was … challenging.

What was the first Linux distro you installed and what year was it?

pgvector is insane. i don't understand why vector databases exist.

Had to stack up 8 Mac Minis to get it running. ~5 tok/sec for now. First time running inference on 8 Mac Minis - performance can be improved a lot (theoretical limit is >10 tok/sec on this setup).

When SaaS products grow and the architecture starts to break down, two types of devs emerge: 1. Those who believe the system is beyond repair and needs to be totally rebuilt 2. Those who put their heads down and fix it one piece at a time I've only seen group #2 be successful.

People have too inflated sense of what it means to "ask an AI" about something. The AI are language models trained basically by imitation on data from human labelers. Instead of the mysticism of "asking an AI", think of it more as "asking the average data labeler" on the internet. Few caveats apply because e.g. in many domains (e.g. code, math, creative writing) the companies hire skilled data labelers (so think of it as asking them instead), and this is not 100% true when reinforcement learning is involved, though I have an earlier rant on how RLHF is just barely RL, and "actual RL" is still too early and/or constrained to domains that offer easy reward functions (math etc.). But roughly speaking (and today), you're not asking some magical AI. You're asking a human data labeler. Whose average essence was lossily distilled into statistical token tumblers that are LLMs. This can still be super useful ofc ourse. Post triggered by someone suggesting we ask an AI how to run the government etc. TLDR you're not asking an AI, you're asking some mashup spirit of its average data labeler.

@mitsuhiko I have a Trollberget in my office as a fallback for my standing desk. It's a good quality stool, durable and functional. If the look works for you, I'd go for it.

We’re in need of new bar chairs since the kids spoiled the old ones badly. I’m struggling to find ones that I like and seriously considering to abuse a bunch of IKEA TROLLBERGET for that purpose. Thoughts?