Projects

Open source infrastructure and tooling built at tiny compute research.

Minimal Multi-Mode Parallel LLM Inference Server

Lightweight LLM inference server with pipeline parallelism, tensor parallelism, hybrid TP+PP, and a dual-backend KV cache that auto-selects FlashInfer or contiguous SDPA based on GPU architecture.

Local Sidecar for Real-Time AI Agent Security

Local sidecar for real-time AI agent security. Monitors agent tool calls via Mamba-2 SSM, detects prompt injection, and blocks attacks before they execute.