SF Big Analytics

Join us on January 7-9, 2026 for an in-person (remote for folks who can’t travel) hackathon at Carnegie Mellon University in Pittsburgh PA, hosted by CMU Libraries in partnership with NVIDIA.

This hackathon focuses on leveraging NVIDIA FLARE to enable collaborative, privacy-preserving computation across multiple biobanks for disease subtyping and genetic analysis, with an emphasis on rare diseases, cancer, and polygenic traits.

Open to graduate students and research scientists in academia, government, and industry. Please complete this brief application form to participate.

💻 https://lnkd.in/e6vRQrGC

A first round of successful applicants will be notified by November 15. Any later applications are due by December 7.

Important: register on the external event website is REQUIRED for admission.

Description:
Welcome to the AI meetup in San Francisco. Join us for deep dive tech talks on AI, GenAI, LLMs and Agentic AI, hands-on experiences with code labs and workshops, and networking with speakers and fellow AI developers, builders, startup founders.

Speakers/Topics:
Check the event website for speakers and topics.

If you have a keen interest in speaking to our community, we invite you to submit topics for consideration: Submit Topics

Sponsors:
We are actively seeking sponsors to support our community. Whether it is by offering venue spaces, providing food/drink, or cash sponsorship. Sponsors will not only speak at the meetups, receive prominent recognition, but also gain exposure to our extensive membership base of 50,000+ AI developers in San Francisco and 500K+ in global.

Registration Link:
Microsoft Teams
Join the meeting now
Meeting ID: 295 116 261 691 43
Passcode: 5vu9QG9o

***

Dial in by phone
+1 949-570-1120,,584919069# United States, Irvine
Find a local number
Phone conference ID: 584 919 069#

Agenda
Talk 1 9:05 am — 9:45 am PT Talk + Q&A
Talk 2 9:45 am — 10:25 am PT talk + Q&A
10:30 am closing

Talk 1: Federated Learning Made Easy: NVIDIA FLARE's Simplified API Stack

NVIDIA FLARE makes federated learning more accessible through three APIs that simplify the developer experience—from adapting existing training code, to building full FL jobs, to experimenting with new algorithms.

The Client API enables developers to transform standard deep learning training code into federated learning client code with as few as 4 lines of code changes. This means your existing PyTorch, TensorFlow, or other ML training scripts can participate in federated learning with minimal modifications.

The JobRecipe API allows data scientists to construct complete federated learning jobs in just 6+ lines of Python code. The same recipe runs seamlessly across simulation, proof-of-concept, and production environments, eliminating the friction of moving from experimentation to deployment. Combined with the Client API, you can go from standard training code to a fully operational federated learning experiment in approximately 10 lines of code.

The Collab API (New!) takes FL algorithm research development to a new level by making the framework essentially invisible. Researchers can develop federated learning algorithms using pure Python function calls—no framework-specific constructs, no steep learning curve. It simply feels like writing standard Python code, enabling researchers to focus entirely on algorithmic innovation.

In this talk, we'll demonstrate how these three APIs work together to transform the federated learning development experience through live examples, showing how barriers to entry have been eliminated for practitioners, data scientists, and researchers alike.

Speaker -- Holger Roth -- Principal Federated Learning Researcher, NVIDIA

Holger Roth, a Principal Federated Learning Scientist at NVIDIA, specializes in developing distributed and collaborative software and models for various industries using federated learning and analytics. He has been exploring the topic both from theoretical and practical standpoints. During the COVID-19 pandemic, he led the experimentation of a federated learning study involving twenty hospitals around the globe to train more generalizable models for predicting clinical outcomes in symptomatic patients. His other research interests include computer-assisted annotation, active learning, and natural language processing. He is an Associate Editor for IEEE Transactions of Medical Imaging and holds a Ph.D. from University College London, UK. In 2018, he was awarded the MICCAI Young Scientist Publication Impact Award.

Talk 2: Building and Deploying Federated Agents Across Multicloud and Edge AI

As AI systems evolve from single models into coordinated collections of autonomous agents, centralized architectures increasingly fail to meet real-world requirements around data locality, regulatory compliance, latency, and heterogeneous infrastructure. This technical, demo-driven webinar presents a practical blueprint for building and operating federated AI agents across public cloud, private infrastructure, and edge environments using the Rhino Federated Computing Platform and NVIDIA GPU-accelerated technologies. Through a live end-to-end demonstration, attendees will see how policy-driven agent placement, secure coordination, local LLM-powered data harmonization, federated cohort construction, and data quality validation can be executed without centralized data movement. The session focuses on concrete implementation patterns, operational considerations, and architectural tradeoffs, equipping ML engineers, platform teams, and AI architects with actionable guidance for deploying federated agent systems in production.

Speaker: Adrish Sannyasi -- Vice President, Customer Solutions and Delivery, Rhino Federated Computing Platform https://www.rhinofcp.com/
https://www.linkedin.com/in/adrish/

Adrish Sannyasi is an accomplished AI solutions leader with deep expertise in cloud data platforms, artificial intelligence, and healthcare and life sciences applications. He has consistently leveraged data and AI technologies to address complex industry challenges and drive measurable business outcomes across the healthcare and life sciences ecosystem. As Vice President of Customer Solutions and Delivery at Rhino Federated Computing Platform, Adrish helps organizations advance their AI initiatives across a wide range of projects, including large language models, protein language modeling, molecular property prediction, healthcare data analytics, EHR data harmonization, and medical imaging AI. Adrish holds a Bachelor’s degree in Electrical Engineering from Visvesvaraya National Institute of Technology (India), an MBA from the University of Maryland, and a Graduate Certificate in Biomedical Data Science from the Stanford School of Medicine.