Pinned Loading

1. llvm-pass-lens llvm-pass-lens Public

   LoRA fine-tuned a 120B LLM to classify LLVM compiler pass interactions. 53.8% → 82.1% accuracy.

   Python

2. bitnet-engine bitnet-engine Public

   Quantisation-native LLM inference engine for 1-bit and ternary models. All matrix ops run via XNOR+POPCNT, never dequantized.

   Rust

3. zerodp zerodp Public

   From-scratch PyTorch DDP and FSDP-style ZeRO-3 sharding, built directly on raw torch.distributed collectives.

   Python

4. Axon-Search Axon-Search Public

   Hybrid semantic search engine combining BM25 + FAISS retrieval, fused via RRF and cross-encoder reranked, backed by an async politeness-aware crawler.

   Python

5. math-finetune-pipeline math-finetune-pipeline Public

   Fine-tuned Qwen3-8B on math — includes the data leak I found in my own benchmark and the GSM8K regression I caused and fixed.

   Python

6. sled-topo-advisor sled-topo-advisor Public

   NUMA/PCIe topology-aware placement hints for AI workloads on Oxide sleds

   Rust