software engineer in NYC. prev https://xitter.gq/t.co/Bx70AilrPF and https://xitter.gq/t.co/ZTUpgwSh6a. Now @powersetres.
New York
Joined November 2007
- Tweets997
- Following1.2K
- Followers873
- Likes11.7K
I think this has been the big transition of 2026 - people kinda just stopped mentioning “vibe coding” cause that’s just how so much gets built now 😳
not just semantic search or clustering, but for example regression: arxiv.org/abs/2411.14708
I think using embeddings with “traditional ML” is really cool and maybe under explored (or at least under hyped) for unlocking a different and cheaper kind of intelligence
EmbeddingGemma 2 maps text, images, video, audio, and code, running 100% locally on your device.
At just 740M parameters, you can build powerful multimodal apps running entirely in your browser via WebGPU & Transformers.js. From multimodal search to cross-modal classification and clustering, local embeddings unlock endless possibilities.
Patrick DeVivo retweeted
It's been a while since I showed any Jev stuff but I've been cooking on a way to use @GoogleDeepMind's EmbeddingGemma2 as an insanely fast real-time /v1/systemone endpoint (20ms on text, 100ms on images).
It's trainable in real-time, can distill an upstream teacher model, and is multi-modal by default. Hard to believe a @googlegemma model of this size can do so well with the right harness!
very cool. I'm trying some long-running @opencode sessions in goal mode to see what it can do. made a CLI to run them in @sailresearch sandboxes
github.com/powerset-co/longs…
on @PowersetRes we run an LLM embedding pipeline to categorize repos based on their READMEs and website content.
we use embedding similarity to a category centroid to determine membership- it works, but seems like a place Jev could really shine
could simplify a whole lot and probably be faster/cheaper
Patrick DeVivo retweeted
12,000 tok / sec on a $250 board is nuts
Waiting for LLMs to stream text is going to seem so quaint in a few years...
12,000 tok/s on a 250$ FPGA board. It's a small model, and doesn't answer coherently to any question or prompt really. try it out on trylittlelamb.com/ but don't expect a good chat. at least its VERY fast
I remember interviewing back in 2015 thinking it would be edgy and a way to differentiate myself to answer a question: "how would you implement a photo gallery feature?" truthfully by saying I would search stackoverflow for the most suggested "photo gallery" library and just use that...
didn't get the job
we score repo's based on how many community PRs have been (recently) merged @PowersetRes
for example: research.powerset.co/repo-sc…
