Multi Agent Patterns

Engineering directors burn 45 minutes every morning reconstructing a picture five tools could have assembled. Replace the loop: five parallel collectors, one orchestrator, a confidence score, a 90-second RED/AMBER/GREEN brief. Triage out of working memory, into code.

One orchestrator decomposes the question. Subagents work the threads in isolation. Synthesis weighs the evidence. The brief lands in twenty minutes — not because the model is faster, but because the topology stopped wasting wall-clock on serial wait.

Two LangChain agents burned $47K in eleven days. The model worked. The budget math didn't. Multi-agent cost is a heavy-tailed distribution, monitoring is structurally too late, and only synchronous SDK-level enforcement stops the spiral.

Why single-inference cost estimates fail for agentic workflows — the four-component inference multiplier (call count, context accumulation, tool schema overhead, retry tax) with concrete workflow examples and measurement patterns.