Season 2
Agentic Orchestration
Production-grade orchestration of AI agents — frameworks, topologies, and observability tested in the real world.
Episodes
-
S02E01: I Built the Same Agent in 6 Orchestrators
LangGraph vs CrewAI vs AutoGen vs OpenAI Swarm vs Pydantic AI vs LlamaIndex. Same 3-step research agent built in each — DX, cost, reliability scorecard.
-
S02E02: Monolith, Handoff, or Swarm? Three Topologies I Built in Production
Architecture patterns for multi-agent systems — when each shape works and when each one breaks under load.
-
S02E03: Observability for Agents — Langfuse vs LangSmith vs Arize Phoenix vs Helicone
Four observability tools tested with the same agent test harness. Pricing, trace quality, alerting, and what each one actually shows you.