Context Engineering for Citizen Agents: A deep dive into giving user agents the information they need

Join fellow AITEC members Evan Fire and Siddharth Ganesan for an interactive session on transforming a general-purpose AI chatbot into a personalized, role-specific assistant that streamlines daily work, improves consistency, and boosts productivity.

Evan will walk through the anatomy of an effective AI assistant, using a practical five-step framework for building a secure, personalized workspace that grows smarter as you use it—from defining behavioral templates and creating dedicated project workspaces to building a knowledge base with your own documents and refining preferences over time.

Sid will cover platform-agnostic memory management for agents: LLM wikis vs. Mem0, sharing memory across a team, and using temporal graphs to track domain entities—and how these approaches complement Skills. He'll show how his team applies these techniques to create self-evolving, personalized coding assistants that share knowledge between team members behind the scenes—prior user corrections, prompts, PR comments, and app-specific documentation.

During this session, you'll learn:

• How to structure a personalized AI assistant around your role and responsibilities
• Why memory is essential to better agent and AI outcomes
• How to give agents persistent, platform-agnostic memory that improves with every interaction
• How to share that continuously improving memory across agents and teammates, so knowledge compounds instead of staying siloed
• A step-by-step approach to building and refining your own AI workspace
• Best practices for organizing documents, setting preferences, and continuously improving your assistant
• Real-world examples of how personalized AI can increase productivity and decision quality