platoseed
Building reinforcement learning environments
We're creating reinforcement learning environments for training frontier models.
We're creating the best AI agent for bug finding
Jazzberry is an AI agent that runs code in a secure sandbox to identify bugs directly within GitHub pull requests by cloning the repo into a microVM, executing and inspecting code, and returning a markdown table of findings. It targets developers seeking automated, in-repo bug detection during PR workflow, with a free trial available.
Formerly โProphetโ, โJazzberryโ ยท why startups rename โ

Long-horizon RL environments

Reinforcement learning environments.