▶ Watch the talk
Featured · 2 November 2025
Evals for AI Agents
A practical look at testing agents against the work we actually expect them to do.
Gareth Paul Jones · Product Lead, AI at Poe
From a fuzzy idea
From a fuzzy idea
to an honest
conversation.
That's why I still code. I like the part where research meets a real product — and staying close enough to the work to build its first version.
Watch the thinking in practice ↓
The practice behind the idea.
Selected video talks.
▶ Watch the talk
Making complicated
technology clearer.
My career has moved through AI products, speech and applied ML, developer platforms, and consumer ecosystems. The people using them are the common thread.
Poe / Quora & Zoom
Leading AI product work at Poe since 2024. Previously, research product work on AI Companion, speech, and applied ML at Zoom.
Twilio
Group Product Manager for the API platform, SDKs, and command-line tooling.
Foursquare
Product and engineering lead for the Places API and partner ecosystem.
Earlier chapters
Twitter · Developer Relations Lead · 2013–2016Apple · Product Manager · 2011–2013Google · Associate Product Manager · 2009–2011