Seattle Physical AI Workshop and Meetup - October 20, 2026

Oct 20, 2026

5:30 PM - 8:30 PM PDT

Walker-Ames Room (KNE 225), University of Washington, Seattle, WA

Speakers

About this event

Join us in the Walker-Ames Room at the University of Washington on October 20th for the Seattle Physical AI Workshop and Meetup, co-presented by Nebius and Voxel51.

The evening kicks off meetup-style with lightning talks from local speakers working across AI, machine learning, and computer vision, followed by a hands-on workshop on building physical AI applications with FiftyOne, from curating robotics and sensor datasets to evaluating vision models on real-world data.

A laptop is required to participate in the hands-on workshop - please bring one. Space is limited, so register early.

Schedule

Hosts

Featured Speaker

When 99% Accuracy Isn't Good Enough: The Long (Fish)Tail of Real-World Computer Vision

Real-world computer vision rarely fails on the examples you expect, it fails on the long tail of strange cases you didn’t know existed.

At Ark CV, we build computer vision systems for fish monitoring, where models that achieve 99%+ accuracy can suddenly encounter folded salmon, bouncing lamprey, unusual lighting, or individual fish that cross the camera view hundreds of times.

This talk shares lessons from deploying computer vision in these messy real-world environments and why benchmark accuracy alone isn’t enough. We'll explore how human review, targeted dataset development, and continuous feedback loops can turn unexpected failures into increasingly robust production systems.

Featured Speaker

Mapping Agricultural Fields of the World with RasterFlow

Agriculture depends on timely, reliable insight into what’s happening on the ground: what’s being planted, what’s being harvested, and how fields evolve over time. The Fields of The World (FTW) project was created to support exactly this mission, by building a fully open ecosystem of labeled data, software, standards and models to create a reliable global map of agricultural field boundaries using AI and Earth Observation (EO) data.

Contributing to this effort, Wherobots processed petabytes of satellite imagery to create seamless global imagery mosaics and produce global maps at 10 meter resolution of agricultural boundaries. In this presentation we'll cover how Wherobots RasterFlow works to scale models like Fields of the World and the applications it unlocks.

Networking and Meetup

5:30–6:30 PM

Networking, food, drinks, and lightning talks

Workshop Agenda

This hands-on session uses DROID, a real-world robotics dataset loaded into FiftyOne as a native multimodal MCAP recording, and YOLO11n, fine-tuned live during the session.

6:30–7:30 PM

Welcome + framing: from raw robot logs to a trained detector
Explore a real DROID robotics recording in FiftyOne's native multimodal MCAP viewer: camera, proprioception, and language on one synced timeline, no ROS install required
Curate: extract and browse frames from the recording, filter and deduplicate
Compute embeddings on the curated frames; explore via similarity search and embeddings visualization (via Nebius Serverless AI Jobs)

7:30–8:00 PM

Auto-label: open-vocabulary detection to generate bounding boxes for the robot gripper and target objects
Train: fine-tune a YOLO11n detector on the auto-labeled frames (via Nebius Serverless AI Jobs)
Evaluate results and close the loop: view predictions back on the original MCAP timeline

8:00–8:30 PM

What else Nebius offers: Token Factory walkthrough — chat/vision models, fine-tuning, credits