North introduced Noros, a FinOps agent designed to answer cloud-cost questions in real time and route them through specialized analysis agents.
AI cost control is becoming an interactive workflow, not just a dashboard: teams want questions answered while spend decisions are still changeable.
The same pattern applies to token budgets: let agents explain spend, but require clear ownership, audit trails, and human approval before automation changes budgets.
Treat Noros as a useful FinOps-agent example and a reminder that cost agents need domain-specific constraints, not open-ended autonomy.
langchain-ai/langgraph
A framework for building resilient stateful agents with explicit graphs, persistence, human-in-the-loop flows, and controllable execution.
getzep/zep
A memory layer and integration collection for AI agents and knowledge-graph-backed language-model applications.
BerriAI/litellm
An OpenAI-compatible gateway and SDK for calling many model providers with budgets, logging, load balancing, guardrails, and cost tracking.
A build-focused walkthrough of a multi-agent FinOps control plane: rule-based triggers plus LLM reasoning to recommend cloud cost actions, with a UI and human approval in the loop.
Augment Code rounds up model routing options for agent systems - tools that decide which model to call per step to balance quality, latency, and cost.
Augment Code breaks down why adding agents can explode costs: orchestration overhead, context handoffs, retries, and verification loops often dominate raw model pricing.