18 minutesPydantic is all you need: Jason Liu
Learn why schema-first LLM calls need typed objects, validators, retries and explicit handling for malformed or hallucinated fields.
AdvancedAI for Business
Andrej Karpathy: Software Is Changing (Again)
Y Combinator. Karpathy's AI Startup School keynote frames LLMs as a new kind of computer — utility, fab, and OS rolled together — and argues for "partial autonomy" products with a human-controlled leash. It is the cleanest articulation of the stack-level mental model the article assumes: that you are picking inference vendors and tooling for a programmable substrate, not a chatbot.
AI Expert note
Model names, pricing and capabilities change quickly. Use this for the decision pattern, then verify current model behavior before adopting it.
What you should get from this
Evaluate partial-autonomy product ideas where humans keep control over high-risk decisions.
Watch or know first
None — watchable cold; it pairs well after the deep dive's training-recipe grounding.
Last reviewed: May 18, 2026
Watch next
Continue through the same learning path with the next curated companion videos.
18 minutes
77 minutesAI prompt engineering: A deep dive
Hear how production prompt engineers revise instructions, examples and behavioral constraints under real pressure instead of treating prompts as one-off text.
AdvancedPrompt Engineering
55 minutesHow to Systematically Setup LLM Evals (Metrics, Unit Tests, LLM-as-a-Judge)
Design an eval ladder that catches regressions before prompt or model changes reach users.
AdvancedAI for Business
Take it further
Hand-picked external courses that go deeper on this topic.
Coursera · DeepLearning.AI + AWSGenerative AI with Large Language Models
Antje Barth · Shelbee Eigenbrode · Mike Chambers · Chris Fregly
When practitioners ask "what should I take if I'm serious about building with LLMs?", this is the answer. Mathematically honest without being a research paper; AWS-flavoured deployment chapters stay useful even if you'll never touch SageMaker.
Advanced~16 hours
DeepLearning.AI · LlamaIndexBuilding Agentic RAG with LlamaIndex
Jerry Liu
Naive RAG (vector search → stuff into prompt) is the version that ships first and disappoints fastest. This short course shows you the upgrade: an agent that plans retrieval, picks tools, compares sources, and answers multi-document questions. Two hours, immediately applicable to a real product.
Intermediate~2 hours
DeepLearning.AI · LangChainLong-Term Agentic Memory with LangGraph
Harrison Chase
The context-window-amnesia problem is what makes most agents feel forgetful and dumb on day two. This course teaches the LangMem patterns that fix it — episodic, procedural, and semantic memory — so an agent remembers user preferences and past interactions across sessions without bloating the prompt.
Intermediate~1.5 hours