From Quality to Autonomy: MLflow + NANDA for the Next Wave of AI Agents
Developers are moving from single models to full autonomous systems. Tonight, we focus on two pillars:
• MLflow GenAI for evaluation, monitoring, and quality observerability
• NANDA for decentralized coordination, identity, resolution, and open source agent infrastructure
The program combines real examples and demos, open discussion, and practical takeaways for anyone experimenting with agent frameworks or planning their next agent-powered application.
6:00 – 6:30 PM
Arrival, Registration & Check-In (Managed by Databricks)
Lite bites + beverages served throughout.
open source demo screens rolling: MLflow evaluation plots, NEST visualizations, AgentFacts workflow, and join the community slides
6:30 – 6:40 PM
Welcome & Opening Context
Speaker: Jules Damji (Databricks)
Framing theme: Quality → Coordination → Autonomy.
Why developers need both MLflow and NANDA to ship the next generation of agent apps.
6:40 PM - 7:10 PM
Session 1: Build High‑Quality AI Agents Faster with MLflow
Speaker: Danny Chiao
LinkedIn Profile: https://www.linkedin.com/in/danny-chiao/
Talk Abstract: One of the top challenges in building an agent is ensuring high-quality outputs. Today, this involves labeling and analyzing traces by hand and iterating on the agent code. In this talk, you’ll learn how to use MLflow to accelerate this process and quickly build a high-quality agent, leveraging techniques used by leading companies to deploy agents in production.
Speaker Bio: Danny Chiao is an engineering lead at Databricks, leading efforts around data observability (data quality, data classification) and agent quality. Previously, Danny led efforts at Tecton (+ Feast, an open source feature store) and Google to build ML infrastructure and high-scale ML-powered features. Danny holds a Bachelor’s Degree in Computer Science from MIT
7:15 – 7:35 PM
Session 2 — Project NANDA: From Registry to Resolution
Speaker: Ramesh Raskar (MIT / NANDA Founder)
Insightful overview of NANDA the open-source architecture and roadmap powering the Internet of AI Agents.
Dr. Ramesh Raskar is an Associate Professor at the MIT Media Lab and Director of the NANDA program, where he leads efforts to architect the "Internet of AI Agents" and decentralized intelligence infrastructure. Previously, he pioneered computational imaging breakthroughs like femto-photography and founded public-interest initiatives including PathCheck Foundation and REDX. A recipient of the Lemelson–MIT Prize and ACM SIGGRAPH Achievement Award, Ramesh holds a PhD from the University of North Carolina at Chapel Hill.
LinkedIn: https://www.linkedin.com/in/raskar/
7:35 – 7:50 PM
Session 3 — Technical Deep Dive: NEST
Speaker: Maria Gorskikh/Karan Bhardwaj
A developer-centric architecture walkthrough:
• Join39.org — the technical core powering agent identity and verified AgentFacts
• NEST.projectnanda.org — sandbox + testbed for multi-agent experiments
• NANDA Adapter — next-gen agent routing (beyond DNS)
⚡ NEST Snapshot
NEST (Network Sandbox & Testbed) is Project NANDA’s unified environment for building, testing, and verifying autonomous AI agents at planetary scale. It’s where decentralized intelligence becomes measurable — a live, visual network connecting agents, registries, and clouds across the globe.
Inside NEST, developers and researchers can:
Launch & Monitor agent networks across regions (US, EU, APAC).
Simulate real-world interactions through safe, policy-aware sandboxes.
Benchmark ROI & Reliability for multi-agent systems in action.
Observe Global Flows of requests, responses, and trust across the NANDA Index.
NEST transforms abstract ideas like agentic coordination, policy routing, and sovereign AI into living, testable systems. It’s not just a dashboard — it’s a working map of the Internet of Agents.
Hosted by Project NANDA - SF Bay Area Chapter, this event brings together researchers, founders, and builders to explore the emerging landscape of Agentic Web— where autonomous AI agents transact, negotiate, and collaborate across digital economies.
7:55 PM – 8:05 PM
Open Q&A
8:05 – 8:25 PM
Community Networking
8:25 – 8:30 PM
Closing Remarks & Community Photo
Final takeaways, contribution paths, and next meetup announcement.
About NANDA
NANDA is pioneering the foundational infrastructure for the Internet of AI Agents — enabling autonomous systems to interact, transact, and collaborate seamlessly.
NANDA employs a three-pronged approach: (i) Technology: Building technology, standards, and reference implementations through a collective of top minds. (ii) Social Mission: Working to maintain an open Agentic Web for innovators while ensuring safety for vulnerable populations. (iii) Venture Ecosystem: Fostering a wide coalition of founders and investors.
NANDA's development draws on proven MIT research in AutoML, split learning, privacy-preserving architectures, and decentralized coordination. Early partners in the coalition gain preferential access to emerging standards, reference implementations, and the academic-industry consortium that will influence the emerging frameworks globally.