Karpathy's Stanford AI engineering lecture: from LLM to Graph in five steps
irinarish · x · 2026-09-20
A widely shared recommendation of Andrej Karpathy's 1-hour Stanford lecture on AI engineering, praised as one of the best explanations of how AI systems actually work. It lays out a progression: 10% LLM → 30% Prompt → 50% Agent → 70% Loop → 100% Graph.
The key takeaway: AI engineering isn't just about writing better prompts — it's about building systems around models with context, memory, tools, feedback loops, and data flows. Summed up as "Delete everything, keep Graph," and recommended for anyone building AI agents.
More from coding & agent
- Devs debate running stateful AI agent runtimes on Cloudflare Workers and other edge runtimes — merlinofthewater · 2026-09-20
- Jev's instant compression scores each tool call to trim agent context without summarization — FinanceYF5 · 2026-09-20
- Stop using LLMs for ticket triage: Jev turns unstructured input into an executable smart if — sven_ai · 2026-09-20
- Jeff Dean's 1-Hour AI Engineering Lecture: From LLM Basics to Agent Graphs — irinarish · 2026-09-20
- Multi-agent systems work best with clear roles, not more agents — _jaydeepkarale · 2026-09-20
- Jev founder: all AI models are built for human-in-the-loop, not true automation — hardimanjames · 2026-09-20