Most people can build a chatbot, but few can build an autonomous agent. What is the secret sauce that allows a probabilistic model to execute deterministic functions?
Join us for a technical deep dive into the mechanics of Tool and Function Calling. We will peel back the curtain on how LLMs transition from mere text predictors to active agents. We'll explore the engineering behind decision-making: How does a model recognize a tool's necessity? How do we enforce structured JSON outputs without hallucination? And how do we manage the 'reasoning loop' where the model interprets a tool's output to decide its next move?
We will go beyond the high-level APIs to discuss the underlying prompt engineering, parsing logic, and state management required to build reliable systems. We’ll wrap up with a live, hands-on demo where we build a functioning agent that interacts with real-world tools to solve a complex, multi-step scenario.
Speaker
Ravi Chandra - Associate Technical Architect at Netwoven
About the Speaker
Mr. Ravi Chandra is an Associate Technical Architect at Netwoven, a Dublin, CA-based company specializing in Microsoft product consulting.
Ravi Chandra focuses on designing, delivering, and implementing large-scale enterprise systems for Fortune 500 companies using M365 and Azure Cloud. His areas of expertise include Microsoft SharePoint Online, on-premises to Cloud migrations, SPFx solutions, .NET, JavaScript, React, Cloud Computing, PowerShell, and the Power Platform. With extensive experience in project delivery, he has played key roles in development, infrastructure, project management, and execution across various domains.
Location
1981 Chester Dr, Tracy, CA 95376
Parking and Basic Information
The venue has plenty of no-cost parking. If possible, use the south parking lot, closer to Lowell Ave since it is closer to the classrooms within the building.