AI Engineering
Building reliable systems on models you did not train — prompting, retrieval, agents, evaluation, cost and security
AI Engineering Crash Course
Fifteen animated sections: next-token prediction and tokenization, the context window and the stateless chat contract, sampling, prompting, structured output, tool calling and MCP, context engineering, embeddings, RAG, the agent loop, workflow patterns, evaluation, caching and prompt injection.
Start learning →AI Engineering Detailed Course
62 sections covering transformers and tokenizers, RLHF and reward design, prefill/decode and the KV cache, prompt management, MCP servers, chunking and hybrid retrieval, reranking, GraphRAG and SQL RAG, ReAct and plan-and-execute loops, checkpointing, agent memory, multi-agent deadlocks, LLM-as-judge, adversarial evals, LoRA and QLoRA, vLLM and continuous batching, cost engineering, the OWASP LLM Top 10, voice agents and production gotchas.
Start learning →