
Private beta
Whale Loop
An execution harness that keeps AI coding agents on the goal, and lets them call work done only after a real check passes.
Ask for early access
Colony tactics. A tiny rebellion against the giants of the forest. Nothing to install: play on the web, on a phone held sideways or on a PC.
Our flagship game
Send a squad with a single card and take down the forest's bosses one by one. One quick stage, or all the way to the end of the forest.
Drag a card onto the field and a squad heads out. At a worksite they build, by the cargo they push, by the boss they bite. Where you drop it decides the job.
The spider queen's silk, the mantis's double strike, the lion's roar. Each boss has its own named attacks and tells to read and dodge.
Start in the browser with nothing to install. It runs on phones held sideways and on PCs, and keeps going offline once added to your home screen.



01Spider
02Spider Queen
03Mantis
04Dragonfly
05Frog
06Snake
07Owl
08Fox
09Eagle
10Wolf
11LionCompany
Games people want to play to the end, and AI agents that see work through to the end. LoopRain is an early-stage startup, started by one person in May 2026.
Make it, ship it, carry what we learn into the next version, and repeat steadily like rain. That is what the name means.

AI agents · In development
Instead of stacking another decision engine on top of the model, Whale Loop lets today's model own the work from exploration to verification, and holds the boundaries firm.

Private beta
An execution harness that keeps AI coding agents on the goal, and lets them call work done only after a real check passes.
Ask for early access
In development
Plan, approve, build, and verify with your coding agents in one desktop app, on the Codex, Claude Code, Cursor, and Grok setups you already have.
Ask for early accessEven in long tasks, it checks the current state against the original goal and makes the biggest gap the next move.
Vague requests start by reading the codebase, become a decision-complete design, and get built only once approved.
Not the model saying "done", but a real check run after the last change is what counts as finished.
Out-of-scope changes, external writes, and human decisions are surfaced, not hidden. Existing changes are preserved.
Works with
Starting small, shipping one thing at a time.
Started as a one-person startup making games and AI agents.
An execution harness that has AI coding agents prove completion with real checks.
OngoingA desktop app for working with agents from plan to verification.
OngoingOpened the first beta of AntWave, our first game, on the web.
OngoingAntWave beta feedback, early access to Whale Loop, or collaboration ideas: send it all by email.