Durable Memory for AI Agents: Build Lab with MongoDB
Hosted by AI After 5 + Mina Abdulla + Max Kuivenhoven + Samadnya Kalaskar + MongoDB
Register with organizer
What to look forward to
This three-hour hands-on lab will guide developers and technical founders through building a Python AI agent with LangGraph, tools, vector search, and persistent memory on MongoDB Atlas. Led by MongoDB Senior Solutions Architect Samadnya Kalaskar, the session will also cover agent reasoning patterns and offer a starter kit and verified skill badge.
Inside the event
AI agents need more than just an LLM, they need reasoning, tools, and durable memory that survives restarts and stays accurate across conversations. Without it, agents forget context, repeat themselves and get more expensive with every conversation.
Join MongoDB and AI After 5 for a hands-on lab led by Samadnya Kalaskar, a Sr. Solutions Architect at MongoDB. Following a brief walkthrough of agent architecture, you'll dive straight into a guided lab to build an end-to-end AI agent from scratch.
AI After 5 is a women’s community, but our build labs are open to all.
🏆 Build it, win it
Finish the lab with your agent's memory running on MongoDB Atlas and you're entered into our raffle for AirPods, Meta Ray-Ban glasses or an Oura Ring, plus MongoDB swag. Winners are drawn at the end of the night, so you must be there to win.
What you'll do
- Orchestrate with LangGraph: Structure agent loops, decision nodes, and conditional edges using Python and LangGraph.
- Integrate Tools & Vector Search: Connect your agent to external tools and execute vector queries using MongoDB Atlas Vector Search.
- Implement Durable Memory: Store conversational history and persistent state in MongoDB so your agent remembers context across sessions.
- Explore Agent Reasoning: Experiment with prompt patterns like Chain-of-Thought (CoT), ReAct loops, and reflection.
Who this is for
- Engineers building agents, copilots or chat products
- Technical founders whose product needs to remember its users
- Developers who've built with LLM APIs or RAG and want to add memory
What you leave with
- A fully functioning AI agent running in Python with tools, vector retrieval, and persistent MongoDB memory.
- MongoDB Starter Kit: Ready-to-run agent templates to continue building after the workshop.
- Official Skill Badge: The verified "AI Agents with MongoDB" skill badge for your LinkedIn/portfolio.
Format
A three-hour hands-on lab in New York. Bring a charged laptop. There's no setup because the lab runs in your browser. Food and drinks will be provided.
Space is limited and registration is approval-only.
About the hosts
MongoDB is a developer data platform for building modern applications. MongoDB Atlas unifies operational data, vector search and application state so AI systems can retrieve fresh context and act on it in real time.
AI After 5 is a global community of 500+ women learning, exploring and building with AI, across New York, London and LA. We run hands-on workshops, build days and curated dinners for people who want to actually use AI in their work. aiafter5.com · @aiafter5
New York, NY
Venue listed by the organizer · New York, NY. Confirm the final address before traveling.
Open in Google Maps ↗Details come from the event listing and the organizer page; registration is managed by the organizer.
Last checked: Nov 11, 2026. Confirm current availability before making plans.
Organizer page checked: Nov 11, 2026. Check registration for the latest details.