Machine Learning & AI

Notes on the math, models, and practical techniques behind machine learning and AI — this knowledge base's core specialty. Expect a mix of theory (the math that makes models work) and practice (what actually happens when you build and ship them).

Articles

• Neural Networks: Math Fundamentals — the linear algebra and calculus behind forward passes, loss functions, and backpropagation.
• From NLP to LLMs: the Mathematical and Architectural Path to GPT-3.5 — the chain from TF-IDF and RNNs through attention, Transformers, GPT, scaling, and instruction tuning/RLHF to ChatGPT.
• FastMCP: Building and Deploying an MCP Server — exposing tools to an LLM client with FastMCP, and deploying one to AWS Lambda.
• Amazon Bedrock AgentCore: Production Infrastructure for AI Agents — what AgentCore is, how it differs from LangChain/LangGraph and the LLM provider, and a migration path for an existing agent.

Contributing

Learning something new about ML/AI — a paper, a book chapter, a technique from work? Add it here as its own file. See Contributing for the how-to.