Personal Digital Governance Policy
The document explains how my digital artifacts and platforms should be organized, disclosed, and handled if I die or become incapacitated.
Jifeng Wu's Personal Website
Hugging Face Models: Repositories, Serving, and the Inference Engine Landscape
A practical guide to Hugging Face model repositories: what they contain, how Transformers loads them, and how serving runtimes such as vLLM, llama.cpp, and Ollama fit into the ecosystem.
Managing Development Environments with conda, nvm, and rustup
Modern development often requires more than just installing one language runtime globally and calling it a day. Installing everything system-wide quickly becomes messy, hard to reproduce, and annoying
Building C/C++ Applications in a Conda Environment
PrerequisitesCommon Build VariablesDirectly Used by gcc/clangThese environment variables are automatically used by gcc/clang themselves. CPATH Colon-separated list of directories to search
llama.cpp Workflow: Download, Convert, Quantize, and Serve Models
llama.cpp supports 90+ model architectures through its convert_hf_to_gguf.py conversion script. The architecture is auto-detected from the model’s config.json. Run python convert_hf_to_gguf.py --print