tilde.run
Let AI agents loose on production. Without the risk.
Tilde is a sandboxed execution platform that makes running AI agents on real production data safe and auditable. It composes GitHub, S3, and Google Drive into a single versioned POSIX filesystem, runs agents in isolated containers with network policy enforcement, and treats every run as an atomic transaction that can be rolled back with one command. It is built by the team behind lakeFS.
Tilde mounts code and data sources as a versioned filesystem, runs the agent in an isolated container with network call auditing and RBAC policies, then atomically commits or discards all changes at the end of the run.
AI/ML engineers and platform teams deploying autonomous AI agents against production data
Background.
- Status
- waitlist
- Business model
- freemium
Similar projects.
Editorial take on the space this project sits in — momentum signals, adjacent moves, our call on whether the wedge is real. Get pinged when we publish a new read or when the landscape shifts.
Have a take on this space?
Tell us what you’d build differently, where you think the incumbents miss, or what we’ve gotten wrong about this project. Comments + reactions are coming soon.