Takayoshi Nakamura GomezT. Nakamura Gomez
← All projects

Project 04 / 06 · Jul 2026 · Tracker

Property Watcher

A tracker that reads property listings, scores each one against what I'm looking for, and keeps a file on every place I look at.

A 29-second loop in painted anime style, set in Kansai. Quiet, until a place fits: a phone on a morning table by a window over a hillside street buzzes, and Sky messages: JUMP, House by the river, meets your jump criteria. Read: listing emails arrive in Japanese and AI pulls out the details in English, marked unverified. Check: plain rules decide, listings land as Pass or Watch, and only the Jump sends a message. Filed: one card per place, next step book a visit for Saturday. The card photo opens into the house by the river, then a walk up the path to its gate, impression Promising. Screens are illustrated; places are fictional; scenes are AI-generated.
How a place gets picked, in a Kansai house hunt. Illustrated screens with fictional places; painted scenes are AI-generated.

How it works

  1. 01

    New listings arrive by email, and AI pulls out the details.

  2. 02

    Clear rules score each one against what I'm looking for.

  3. 03

    Sky messages me only when one fits, and every place gets its own file.

Why I built it

Listing alerts are noisy. I wanted to see only the few that fit, and never lose track of a place I’d already looked at.

What I learned

  • AI is good at reading messy listings and bad at making the final call. The scoring stays in plain rules, and every rule is tested.
  • One file per place beats a pile of alerts.
  • Several AI agents can share one record when it’s clear who edits what.
A shared tracker page with the same places, showing which agent last edited each row. Dummy rows.
The shared page Grok Bot keeps in sync. Illustration with dummy rows.

Built with: Python · SQLite · Claude · Sky · Grok Bot · Notion

Follow along

What I learn, as I go.

I share what I built, what broke, and the tech worth knowing about.