LLM Architecture
Attention, KV cache, normalization, positional encoding, and memory mechanisms.
Wenhao Liu
I write technical notes that turn model architecture and infrastructure details into clear engineering references. Current focus: attention mechanisms, LLM memory, PyTorch, Linux, and networking fundamentals.
- Focus
- Agent engineering
- Published notes
- 0 technical write-ups
- Latest update
- Loading...
About
Building a sharper engineering notebook.
This site is my public workspace for studying deep learning systems and writing down the parts that are easy to forget: tensor operations, Transformer components, decoding efficiency, Linux usage, and practical network paths.
I care about explanations that connect math, code, and systems behavior, so each note is written as a reusable reference for future projects and interviews.
Focus Areas
Where I am spending attention.
ML Engineering
PyTorch workflows, Hugging Face tooling, tensor operations, and reproducible experiments.
Systems Thinking
Linux, shell tooling, Docker basics, Git.
Skills
A practical stack for AI engineering.
Programming
ML Frameworks
Systems & Tooling
Projects
Selected project notes.
Writing
Technical notes and study logs.
Loading notes...