Personal AI agent
Jeeves
A personal assistant that organizes my day and recommends things based on what it has learned about my taste.

01
The problem
Everything I needed to organize my day was already available, but it was spread across my inbox, calendar, maps, transport services, and the weather forecast. I still had to read it all and work out what actually needed my attention.
02
The approach
I built Jeeves around a daily brief, then expanded his role as he learned more about me. He brings the relevant information together, applies what he knows about my preferences, and helps me decide how to plan the day, where to eat, and what to do.
03
How it works
Jeeves reviews my calendar and relevant emails, checks locations, travel time, and weather, then produces one concise brief. He also remembers details such as how early I like to arrive, what I consider nearby, and the restaurants, places, and experiences I tend to enjoy. That allows him to recommend things for me, rather than return a generic list of popular options.
04
Key decisions
- Start with one dependable daily brief before expanding Jeeves into a broader personal assistant.
- Treat my taste as core context, alongside practical information from my calendar and inbox.
- Prefer a smaller number of recommendations based on verified details over a longer but less reliable list.
- Store corrections as lasting context so Jeeves does not make the same mistake in a future conversation.
05
Current use
Jeeves is now part of my daily routine. It reduces repeated planning work, catches details that could otherwise surprise me later in the day, and has become the first place I go when I want a recommendation tailored to me.
06
What I’ve learned
Jeeves became personal by remembering small, specific things about me. Corrections to transport preferences, rain checks, addresses, timing, and recommendations all made him better. Knowing what I like has proved as useful as knowing what is on my calendar.