Austin Chen

I like designing stuff
Commerce + CS @queensu

[Austin Chen @austinc808]

@Oriku175 Looks great! how did you make it?

[Austin Chen @austinc808]

@Ahtasham_Design Looks great! Mind breaking down how you created the animation?

[Austin Chen @austinc808]

@seltugba_ Looks great! Is the imagery generated or real pictures?

[Austin Chen @austinc808]

@anirudhbv_ce @huggingface Cool bro

[ani @anirudhbv_ce]

pip install spectralquant ✂️

Up to 6.62x KV cache compression for LLMs and transformers.

Same model. Faster outputs. Smaller KV cache.

Try now (2 mins): https://t.co/7qHRy0DDug

- KV cache integration via @huggingface's DynamicCache
- Three presets: 5.95x (paper), 6.55x (validated), 6.68x (edge)
- Mistral 7B / Qwen 2.5 7B / Llama 3.1 8B verified
- Pure PyTorch + future CUDA kernel support
- Auto-calibration from a bundled corpus

📰 Paper: https://t.co/I89vK1nEFZ

💻 Code, quickstart, and benchmarks: https://t.co/cKJt33vsgk

#LLM #Inference #PyTorch #OpenSource #MachineLearning #LLM #KVCache #Inference

[Austin Chen @austinc808]

@lsnc_1729 @TechTO @UdulaAbeykoon @rootlyhq W win 🥇

[Austin Chen @austinc808]

RT @anirudhbv_ce: This week, we presented STAE-Spectral-Magma at the @RBCBorealis in Toronto.

Our novel hybrid architecture tackles city-s…

[Austin Chen @austinc808]

@anirudhbv_ce Preciate it bro i needed this 💪

[Austin Chen @austinc808]

@ruijianwng I'm not investing in this

[Billy Wang @ruijianwng]

When I was 6, all my classmates wanted to be firefighters, astronauts, and doctors.

I told my teacher I wanted to be an AI-native B2B SaaS founder targeting mid-market with a usage-based
pricing model looking to raise $500k pre-seed.

She called my parents.

[Austin Chen @austinc808]

Dither motion + Japanese typography for a website I'm designing.

Inspired by @Oriku175

Built in Figma

[Austin Chen @austinc808]

Made a math app for kids.

Adults tolerate bad UX. Kids just quit.

Design accordingly.

Created in Figma

[Austin Chen @austinc808]

AI SaaS is cold and robotic.

Nature instills trust.

Built with Figma + Grok

[Austin Chen @austinc808]

@nengjiali Don’t worry that feature will come out in our next update

[Austin Chen @austinc808]

We killed tutorials.

You no longer have to explain how to do things.

CTRL allows you to record any task and send it as an executable link.

You can finally gain back the endless hours your team was wasting.

[Austin Chen @austinc808]

We're building this for teams tired of repeating themselves.

Get early access: https://t.co/E7pJvzBh2q

[Austin Chen @austinc808]

👇Here's how it works:

Someone records a task once with CTRL.

It generates a shareable link.

Whoever receives it can see exactly what will run, approve it, then execute it automatically on their device.

[Austin Chen @austinc808]

@anirudhbv_ce @OpenAI @GeminiApp @sentra_app nice work

[ani @anirudhbv_ce]

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: https://t.co/zut8pdoPbH

[Austin Chen @austinc808]

@UdulaAbeykoon Valid bro 🔥