Building Agents: Designing, Shipping, and Controlling the Cost of AI
Join us for an honest conversation about what it takes to build, deploy, and trust AI agents in production.
This is a technical discussion for engineers and product leaders who are actively building agentic systems. Our panelists have shipped agents across different domains, from cloud cost management to security remediation to foundation model infrastructure, and will share the architecture decisions, hard problems, and real results along the way.
What we'll cover
AI agents are entering production faster than most teams anticipated. Building them well means solving hard problems across design, deployment, and ongoing operations, including getting a handle on what they actually cost to run.
Designing the agent: How production agents are structured today: tools, memory, model integration, orchestration, and decision loops. The patterns that held up, the ones that didn't, and what our panelists would do differently. We'll get into specific platform decisions, including building on AWS Bedrock.
Leveraging AI in engineering practice: How teams are using AI across development workflows, the hard problems worth talking about (anomaly detection in noisy data, balancing safe versus autonomous actions, integrating with live systems), and what it takes to earn trust in an agent that touches real infrastructure.
Controlling and understanding cost: What production agents actually cost across compute, inference, and orchestration, and how teams are treating cost as a first-class engineering concern rather than an afterthought.
The forward look: Where agentic AI is heading, how the economics are shifting, and what engineers building today should be thinking about for what comes next.
Program
🚪5:30 - 6:00 PM: Doors open & check-in.
🥂6:00 - 6:45 PM: Welcome reception with food and refreshments.
🎙️6:45 - 7:45 PM: Panel discussion with AI experts
🗣️7:45 - 8:30 PM: Stick around to meet the speakers and continue the conversation.
We ask all approved guests to arrive between 5:30 pm and 6:30 pm. Everyone must register, no plus-ones.
Panelists
Crystal Marion, Sr. Software Engineer, AI @ North.cloud
Josh Kotrous, CTO @ Pensar
Raj Sakhuja, Generative AI Specialist @ AWS
TBD
North.cloud helps engineering, finance, and cloud teams optimize cloud spend with greater control and automation. The AI-powered FinOps platform delivers automated commitment management, AI-driven rightsizing, real-time cost intelligence, and GenAI workload visibility, helping teams improve commitment efficiency, reduce on-demand exposure, and take control of AI costs.