Hi, I’m Wei. By day, I’m a Senior Staff SWE focused on AI/ML for payment risk—tackling everything from fraud detection to credit modeling.
I use this space to log my learning notes, grounded in the belief that articulation is the ultimate form of reinforcement learning. With the rise of generative AI, I’m spending less time on the syntax of writing and more time on the substance of the ideas.
Posts
-
Multi-Agent Systems in Practice
-
The Evolving Agent: Experience-Layer Learning
-
AI-Native System: From Model to AI Agent
-
vLLM - Revisit
-
Knowledge Distillation - Revisit
-
MTP & MTP-D Deep Dive: Beyond Next-Token Prediction
-
LLM Inference Optimization: 2026 Update
-
Large Transformer Model - Inference Optimization
subscribe via RSS