Krish Chhajer

LFG! Talos on chip cc @luthiraabeykoon

@lsnc_1729 @TechTO @UdulaAbeykoon @rootlyhq yoooo congrats! @UdulaAbeykoon @lsnc_1729

Yooo first place at the @TechTO hackathon w @UdulaAbeykoon Thanks @rootlyhq for the sick venue!!

@anirudhbv_ce @sentra_app @GoogleResearch cool stuff!!

@sentra_app just killed @GoogleResearch's TurboQuant. Introducing SpectralQuant — 5.95× KV cache compression on Mistral 7B at +7.5% perplexity overhead. TurboQuant at the same compression: +22%. 3× less degradation. 15-second calibration. One per-model, then drop-in for any HuggingFace LLM, ViT, ESM, AlphaFold Evoformer, or VideoMAE. 📰 Paper & results: github.com/Dynamis-Labs/s… Check out the findings and how the mechanism works below. ↓

@KanoiKrishnav to the moon 🚀

Career update : joined a startup as founders office intern.

@anirudhbv_ce @OpenAI @GeminiApp @sentra_app the beaver is cooking again really cool read!

We finally know why LLMs hallucinate. It's not the model. It's the geometry. @OpenAI text-embedding-3-large: 91/3072 dimensions do real work. @GeminiApp gemini-embedding-001: 80/3072 dimensions do real work. ~97% of your vector database is mathematically empty. Your RAG system is retrieving from noise. @ashwingop and I present "The Geometry of Consolidation" - a proof that RAG compression has a hard floor no algorithm can beat, set by a single spectral number your embedding model cannot escape. Every hallucination your RAG pipeline produces? This is why. Paper + results: github.com/niashwin/geome…

reinventing Groq's LPU with @michael_trbo we got instruction driven data movement working between SRAM memory blocks and MXM compute!!

what a run with @luthiraabeykoon, can't wait to get started on V3👀 thanks to everyone for the support!

Featured on @AlphaSignalAI, thanks everyone for all the support!

@UdulaAbeykoon to the moon🔥

Introducing CTRL! We just turned any digital task into a single shareable link that runs itself on anyone’s device, automatically. No more docs. No more Loom videos. No more “can you just screen-share real quick?” One click. Done. On any computer.

@freemanjiangg @luthiraabeykoon @karpathy 🫡

@luthiraabeykoon @karpathy canada stays on top

We implemented @karpathy 's MicroGPT fully on FPGA fabric. No GPU. No PyTorch. No CPU inference loop. Just a transformer burned into hardware, generating 50,000+ tokens/sec. The model is small, but the idea is not: inference does not have to live only in software 👇

Final day. We don’t pitch. We deploy.

@KanoiKrishnav @ShreyB007_ Lets gooo🚀

@sakshambatraa @luthiraabeykoon @karpathy 🫡

@luthiraabeykoon @karpathy dynamic duo

@anirudhbv_ce @luthiraabeykoon @karpathy @Rafael_L_Spring ☠️

@luthiraabeykoon @karpathy @Rafael_L_Spring would appreciate your take on this implementation

@luthiraabeykoon @karpathy Another run with the 🐐 @luthiraabeykoon !!

with Mira, AI can now live on your face. capture every conversation. create the most personalized form of AI ever. order now.

@kennykgguo @luthiraabeykoon 🙌🙌

reflections yadayada open.substack.com/pub/kennyguo/p…