Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn practical FinOps strategies to control and optimize costs when deploying AI agents in production environments through this 31-minute conference talk. Discover how AI agents, while appearing intelligent, lack cost awareness and can generate unexpected expenses through poorly bounded retries, excessive API calls, and unrestricted tool usage that can result in hundreds or thousands of dollars in compute, token, and storage costs. Explore real-world lessons from deploying agentic systems in cloud-native pipelines and infrastructure security tools, including insights from DockSec, an open-source AI-powered container security analyzer. Understand how seemingly harmless design decisions like overly verbose prompts, excessive tool chaining, and unrestricted LLM usage lead to runaway spending, and examine how agents misbehave in cloud billing terms. Master practical strategies to monitor, contain, and optimize agent costs by integrating cost observability into your agent stack, programmatically setting retry, token, and API call budgets, and leveraging agent memory, caching, and behavior throttling to reduce waste. Gain actionable tools to design agent systems that operate intelligently while maintaining financial sustainability, whether you're scaling agents in production or just beginning to build them.
