Join us for an exciting talk night at Github's Bellevue office!
Agenda:
5:30-6pm: Doors open/networking
6-6:10pm: Opening remarks
6:10-6:35pm: Talk #1
6:35-6:50pm: Intermission
6:50-7:05pm: Lightning talk
7:05-7:30pm: Talk #2
7:30-7:40pm: Closing remarks
After party to follow at Lucky Strike Bellevue (700 Bellevue Way NE Suite #250, Bellevue, WA 98004)
What we'll do:
This event will feature the following presentations:
Talk #1: Building TowleVision: A Local-First AI Media Production System in Python
Speaker: Tyrell Towle
Description: TowleVision is a local-first, model-agnostic AI media production system built primarily in Python. This talk explores the engineering lessons behind coordinating narration, image generation, captioning, rendering, review, repair, and publishing as one structured workflow. Using sanitized interface views and finished-media examples, Tyrell R. Towle, PhD, will discuss deterministic orchestration around probabilistic models, interchangeable local AI backends, artifact logging, targeted reruns, and the practical strengths and risks of building substantial software with AI coding tools.
Lightning talk: TBD
Speaker: TBD
Description: TBD
Talk #2: Aging in Pandas: Building a Biological Age Clock from Public Health Data
Speaker: Ramkumar Hariharan
Description: "How old are you?" Simple question. You know the answer instantly — it's on your driver's license, it's the number people ask at parties. But here's the thing: that number is basically useless for telling you how healthy you are. I want you to picture two 50-year-olds. Same birthday, same age on paper. One of them has the blood chemistry of a 35-year-old — low inflammation, great metabolic markers, organs functioning beautifully. The other one's blood chemistry looks like a 65-year-old's. Same chronological age. Wildly different biological age. And it turns out biological age — not the number on your ID — is what actually predicts who gets sick, and who doesn't. Today I want to show y