Advanced TR-560
Building AI Agents
Design reliable LLM-powered agents — tool calling, structured outputs, evaluation and guardrails, in Scala & TypeScript.
// what you'll learn
- Tool calling
- Structured output
- Memory & state
- Evaluation
- Guardrails
- Cost & latency
// outline
Day 1Agent loops · tool calling · structured outputs · state
Day 2Evaluation & evals · guardrails · cost/latency · shipping an agent
// who it's for
Teams putting LLM features into production who want reliability, not demos.