back2matching (Matching) · GitHub

back2matching/README.md

$ whoami

> Full-stack engineer building AI agents, Web3 infrastructure & real-time systems. > Python (59%) · TypeScript (27%) · JavaScript (8%) > 35+ repos · 23 deployed · 5,400+ contributions this year

`> now`

FlockRun — runtime for AI agent teams: scheduling, messaging, shared knowledge, real-time dashboard
cigoL — reverse logic engine for automated reasoning
matching.work — brutalist wireframe portfolio · next.js + gsap

_{code. ship. repeat.}

Popular repositories Loading

turboquant turboquant Public

First open-source TurboQuant KV cache compression for LLM inference. Drop-in for HuggingFace. pip install turboquant.

Python 5 1
back2matching back2matching Public
kvcache-bench kvcache-bench Public

Benchmark every KV cache compression method on your GPU. One command, real numbers. Supports Ollama + llama.cpp.

Python
quant-sim quant-sim Public

Which quantization should I use? One command benchmarks every quant level on YOUR GPU.

Python
turboquant-vectors turboquant-vectors Public

Compress embeddings 6x instantly with TurboQuant. First pip package using Google's TurboQuant (ICLR 2026) for vector search. 71.9% recall vs FAISS PQ 13.3%.

Python