In the news
Dynamic workflows for stateful agents and decision models
Letta · Published · 3 min read
In 30 seconds
- What happened
- Letta Code now lets agents write dynamic workflows that spawn and coordinate multiple subagents across different models and decision APIs.
- Why it matters
- Matters for engineers building agentic systems who need to parallelize work, handle large context, or route tasks to cost-effective models.
- Watch out
- Subagents are stateless and receive only what the orchestrator passes them; scaling many subagents quickly multiplies API costs despite per-call savings.
- agent
The patterns behind this
Each one covers how the technique works, when it earns its cost, and where it breaks.
The Agent Architect
One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.
Weekly email, one-click unsubscribe. We only use your address to send the briefing.