I love to build things ✨
Somewhere around the ninth “OpenAI just announced” Slack message in a single week, I gave up trying to hold the timeline in my head. Model names stopped mapping to release dates, release dates stopped mapping to what actually changed, and “the new one” became ambiguous across at least four labs simultaneously. That’s not a personal failing so much as a description of the last two years: the…
Every guard model I’ve used ships with someone else’s taxonomy baked into its weights. LlamaGuard has its hazard categories, ShieldGemma has its harm types, and if your actual policy doesn’t line up cleanly with whatever list the model was trained against, you’re stuck rephrasing your rule until it happens to trip one of theirs. That works fine for the categories the labs anticipated. It works…
On July 6, 2026, Anthropic published A global workspace in language models , based on the paper Verbalizable Representations Form a Global Workspace in Language Models . Within hours the internet had sorted itself into two camps: people who think Claude is now sentient and would like to marry it, and people who think Anthropic discovered that a computer computes. Both camps have one thing in…
Live dashboard: metra.snehal.ai · Code: github.com/spate141/metra-monitor
My wife asked me a simple question: “When should I water the vegetable beds?” A normal person would have said “check if the soil feels dry about an inch down.” I am not a normal person. I know a thing or two about IoT sensors, LLMs, and agents. So naturally, I spent the next two weekends building a complete monitoring pipeline so that neither of us would ever have to go outside and touch grass…
In May 2026, Robinhood shipped an official Model Context Protocol server that lets an AI agent place real trades on your behalf. A few weeks earlier, the SEC scrapped the pattern-day-trading rule, so the old four-trade-a-week constraint is gone. Wire those two together and you can run an LLM agent on a schedule: it wakes up, evaluates a strategy, and executes without a human in the loop.
I wanted to add audio narration to my blog posts on snehal.ai. The idea was simple: drop a text content, get a clean .wav out, ship it with the post. The execution took longer than expected because most open-weight TTS models that sounded good enough were either too large for my RTX 4090, too slow to be practical, or produced artifacts that made long-form narration unusable. I tried a few…
The thing that finally got me to sit down with Ideogram 4 was the text rendering. Most image models treat text in a scene as decorative blur: you ask for a neon sign in Japanese and you get neon-sign-shaped noise. Ideogram has been good at in-image text for a while, and when they open-sourced the weights last week I wanted to know whether “good at text” survived the quantization-to-run-on-one-GPU…
I watched a video this week that I cannot stop thinking about. It is roughly ten minutes, by Mo ( @atmoio ), and on paper it is about how AI works, but it turns into a question about what it means to be human, and it got me. So this is me thinking out loud after watching it. The good ideas here are mostly his. The commas are mine. He opens with a story from 1958. In a basement at Johns Hopkins,…
You’ve got a small student model you want to make smarter. Somewhere out there, a bigger, wiser teacher model has all the answers. Standard knowledge distillation says: have the student match the teacher’s output distribution, position by position, via KL divergence. Elegant. Clean. Works great.