Dead Drop
You carry a small beacon. It senses other players as you move through the city: no GPS, no map, no internet, just proximity. When two agents cross paths the hardware decides the roles: one becomes the hunter, one the hunted. You will never learn their name, only their patterns.
- Role
- Designer & author
- Status
- Paused, on purpose
- Hardware
- BLE beacon
- Built
- 2026
The game
The hunter tries to keep crossing your path. You try to disappear. Each evening a handler debriefs you: enemy agents picked you up around five o'clock. Change your routine. And you realize, with a chill, that you are predictable.
You can plant a hidden device like a dead drop, harvesting intel on everyone who passes and scoring the longer it survives. But to plant it you go dark, and eventually you have to risk retrieving it. Everything you want costs exposure.
Why it is not shipped
The firmware compiles clean and runs the full detect, log and sync loop. I could put it on hardware and hand devices to friends this month. I have not, and that is the entire reason this page exists.
A game about being followed is a tracking system wearing a costume.
I do not yet have an identity scheme I trust to keep players permanently anonymous to each other. Until that scheme is resolved and reviewed by someone who knows more cryptography than I do, it does not go to real people.
The anonymity has to be real, not promised. If a player could work out who their nemesis is (by correlating a persistent identifier with a face they keep seeing on the same train), the game stops being a game. The proposed scheme uses rotating addresses with authenticated bonds. It is written up, and it is explicitly marked as pending review rather than settled.
How the project is built to stop me
Good intentions do not survive a late night and a working prototype, so the guardrails are structural:
- The build is hard-walled so the throwaway test identity cannot ship by accident.
- A privacy and safety document sits at the front of the repository, labelled read before building anything that broadcasts.
- A decision log records every architectural choice, what is settled, and what is still open, so the unresolved item stays visible instead of quietly becoming a habit.
Why I put an unfinished project on my portfolio
Because stopping was the work. It is straightforward to ship something and deal with the consequences later; the harder discipline is recognizing that the interesting technical problem and the responsible thing to do have come apart, and choosing the second one while the first is sitting right there, compiled and ready.
If you are hiring someone to work on systems that touch the public, I would rather show you that judgment than one more finished app.
Honest limits
- It has never run on real hardware with real players. Awaiting first flash.
- No automated tests.
- The anonymity scheme is a proposal, not a reviewed design, and I am not qualified to sign off on it alone.
- Most of the value in this repository is documentation, not code.
Seeing the code
The repository is private. If you are evaluating me and want to read the code or walk through the architecture, ask me and I will arrange read access or a screen share.