Explainers

Agentic AI

Illustrated Coding Agent

Read series →

A visual deep-dive into how coding agents work: the loop, tools, context, orchestration, verification, and failure containment.

Sample chapters

  • The agentic loop
  • Tools and permissions architecture
  • Context engineering and memory
  • Subagents and orchestration patterns
  • Verification and self-correction

Infrastructure

Illustrated LLM Inference

Read series →

A guide to how inference works in production, from token generation and KV caches to batching, quantization, and serving economics.

Sample chapters

  • Tokens and the decode loop
  • Prefill vs decode
  • KV cache mechanics
  • PagedAttention and memory management
  • Quantization strategies

MLOps

Statistics for MLOps

Read series →

A story-driven treatment of the statistics behind production ML: drift, testing, monitoring, and continuous safeguards.

Sample chapters

  • Baseline distributions and drift detection
  • Covariate drift
  • Concept drift
  • A/B testing and SRM detection
  • Continuous monitoring

Alignment

Illustrated RLHF Guide

Read series →

An interactive explainer for preference data, reward models, PPO, Constitutional AI, and alignment-oriented training loops.

Sample chapters

  • Why RLHF matters
  • Reward models from preference data
  • Policy optimization with PPO
  • Reward shaping playground
  • Constitutional AI and tool use