← AI Agent Engineering
Core Skills
From workflows to multi-agent systems to cost/latency engineering: learn how to make agents fast and cheap without just reaching for a bigger model.
Unit 1
Multi-Agent Collaboration
Orchestration topologies, where multi-agent fits, and the costs and failure modes behind the gains.
Unit 2
Cost / Latency Engineering
Starting from the two-phase nature of inference, master the engineering levers that save money and time: KV cache, prompt caching, speculative decoding, batching, model routing, and prompt compression.