tensorlake
View on GitHubTensorlake is a serverless runtime for sandboxes and deploying background agentic applications
Python and TypeScript SDKs for creating stateful Firecracker microVM sandboxes and deploying serverless agent applications. Supports isolated code execution, snapshots, sandbox pools, and distributed orchestration.
Use Cases
Run LLM-generated code in isolated sandboxesGive agents persistent execution environmentsDeploy long-running background agent workflowsExecute tools securely for AI agentsCheckpoint and restore agent workspacesScale orchestration tasks with fan-out
Built With
- Language
- Python
- Frameworks
- OpenAI Agents SDK · LangChain · LangGraph · Qdrant
Tags
agent sandboxes · code execution · microVMs · serverless runtime · agent orchestration · background agents · isolated execution · snapshots · sandbox pools · Python SDK · TypeScript SDK