Zechen Zhang
Community@zechenzhangagi · Cambridge, MA
Building the future of AI-human collaborations
Agent Skills by Zechen Zhang
Showing 9 vetted skills indexed across 1 GitHub repositories.
zechenzhangAGIspeculative-decoding
Reduce LLM inference latency using draft models, Medusa heads, and Lookahead Decoding.
zechenzhangAGIlong-context
Extend transformer context windows to 32k–128k+ tokens using RoPE, YaRN, and ALiBi.
zechenzhangAGImoe-training
Train Mixture-of-Experts models with DeepSpeed MoE and HuggingFace Transformers.
zechenzhangAGImodel-pruning
Prune large language models using one-shot methods like Wanda and SparseGPT.
zechenzhangAGImodel-merging
Merge fine-tuned models using linear, SLERP, task_arithmetic, TIES, and DARE methods.
zechenzhangAGIknowledge-distillation
Compress large language models by transferring teacher knowledge to smaller student models.
zechenzhangAGIpinecone
Manage a serverless vector database with hybrid search and metadata filtering.
zechenzhangAGIchroma
Store embeddings and metadata for semantic search and RAG pipelines.
zechenzhangAGIfaiss
Perform billion-scale k-NN similarity search with FAISS index types and GPU acceleration.