Each pillar ships as a deep, runnable module — first-principles derivations, production tradeoffs, live demos, and a leveled Q&A bank. Depth-first: one polished pillar per session.
New to ML? Start here. The Stanford/CMU core — how models learn, the math you need, neural nets, the road to LLMs — concise, for software engineers.
Enter →ReAct, planner-executor, tool use, memory, multi-agent and MCP — build a Claude-Code/Codex-style coding agent end to end.
Enter →Chunking, hybrid search, rerankers, ColBERT, GraphRAG, contextual retrieval, agentic RAG — and how to evaluate it.
Enter →RLHF/RLVR, PPO → GRPO and the variant zoo (DAPO, GSPO, Dr.GRPO…), RL infrastructure, and the 35-question RL interview benchmark, answered.
Enter →Chaining, routing, parallelization, orchestrator-workers, evaluator-optimizer; workflows-vs-agents and durable execution.
Preview →vLLM, PagedAttention, KV cache, speculative decoding, quantization (AWQ/FP8), FSDP & tensor/pipeline parallelism.
Preview →LLM-as-judge, RAGAS/DeepEval, golden datasets, agent trajectory evals and CI regression suites.
Preview →Attention/MHA, RoPE, RMSNorm, activations, tokenization, sampling, and scaling laws — from first principles.
Preview →The context-engineering paradigm, prompt caching, long-context, lost-in-the-middle, structured outputs, CoT.
Preview →CopilotKit/AG-UI, generative UI, human-in-the-loop; harness engineering for production coding agents.
Preview →Design a RAG bot over 10M docs, an AI coding agent, an eval platform, a voice assistant, an agentic research system.
Preview →Constitutional AI, RLHF→RLAIF, prompt-injection/jailbreak defense, the 6-layer guardrail stack, red-teaming.
Preview →Leveling IC3→staff, OpenAI/Anthropic loops, coding rounds, behavioral/values, and the cross-pillar Q&A bank.
Preview →