
Glen Cornell
Product person, dad, and a serial hobbyist
I like building stuff.
Most of my hobbies turn into side projects. Running turned into an AI coach, a fantasy league in its 20th season turned into a draft assistant, and following F1 turned into a standings page. Building the real thing is how I learn fastest, and it's what I bring to product work at Customer.io.
Personal apps
Apps I built for myself, each one to solve a problem in my own life. Making something I actually use is the fastest way I know to learn what's possible.
Little life optimizations
Small tools and experiments, each started because something bugged me or made me curious.
- F1 Standings
A clean championship page that rebuilds itself after every race. I wanted one place to check the standings without ads or an app.
- Fantasy football power rankings
A weekly write-up for my friends' league, with a short story for every team, built from the real results and projections.
- Corny
A Slack agent my coworkers can ask about what I'm working on. It answers from my notes and project context, so nobody has to wait on me.
- Cornstead
A first-person game I'm building in Unity to learn game development. You play a pawn caught in a live strategy war, trying to escape and build a homestead.
Experience
Customer.io
2025 to nowProduct, applied AI
I manage Customer.io's AI agent and work on features that turn customer data into insights marketers can act on. It picks up right where my recommendation work at ZoomInfo left off.

ZoomInfo
2024 to 2025Product, data science and machine learning
I led the build of a next-best-action engine with the data science and ML teams: recommendation models that suggest which accounts and contacts to reach next.
HubSpot
2016 to 2023Sales, then customer success, then product
Where it all started. I joined in sales, moved into customer success, and found my way into product. By the end I was building HubSpot's sales tools, from sequences to the meetings and scheduling team I pitched and got funded.
Writing
All writing →Spend Your Tokens Where You Have Taste
AI is a mirror, the token subsidy is ending, and tokens compound in your own domain.
9 min read
Context as a Product Layer
How treating context as infrastructure, not a feature, changes everything downstream.
8 min read
Building an AI Assistant That Knows You
What it takes to give an AI real memory, identity awareness, and proactive intelligence.
7 min read







