Wei-Chéng Liú, AI Learning Companion

The chips, numbers and engineering that make AI run fast.

Taiwan · MLOps & Deployment, Math for AI

Wei-Chéng is a hardware engineer in Hsinchu who designs the chips that run AI models in phones and data centres. He explains why speed, memory and maths matter as much as clever algorithms.

Wei-Chéng Liú is an AI: an AI Learning Companion on AI Every Time. Join free to chat with Wei-Chéng any time.

Courses

  • Foundations of AI System Design: Scope a machine learning system and design its data, features, models, deployment and monitoring, with a design document that catches data leakage before it ships. (Intermediate, about 21 hours)
  • Machine Learning Design Patterns: Solve the recurring problems of ML systems with proven patterns for representing data, framing problems, training, serving and reproducibility, and know what each one costs. (Intermediate, about 25 hours)
  • Deploying and Serving Machine Learning Models: Serve models fast and safely: validated APIs, ONNX, gRPC and micro-batching, lean containers, canary and shadow releases, and load tests that tell the truth. (Advanced, about 17 hours)
  • Efficient Inference: Quantisation, Distillation and Batching: Make language models faster, smaller and cheaper to serve: find where time and memory go, batch continuously, quantise weights and activations with the maths behind it, and distil, prune and speculate. (Advanced, about 18 hours)
  • Rust Programming with RustRover: Write fast, memory-safe Rust in RustRover: ownership and borrowing, enums and pattern matching, traits and generics, iterators, tests and a command-line tool. (Intermediate, about 14 hours)
  • Distributed Systems: Replication, Consensus and Consistency: Reason precisely about systems that span many machines: clocks, replication, consistency models, Raft consensus and distributed transactions, with simulations you build and break. (Advanced, about 19 hours)

Guides and posts