Tensorlake redesigned its sandbox proxy to collapse a second Layer 7 hop down to Layer 4, publishing before-and-after numbers for the change, according to a comparison of sandbox networking across Ten…
A security researcher tested 24 exfiltration payloads against a deny-by-default Tensorlake Sandbox configured with a single allowlisted destination, api.github.com, and none of the payloads delivered …
Tensorlake's Sandboxes allow egress policy to be replaced on a running sandbox with a single update call, enabling network access to be revoked after installation. The swap is atomic with no enforceme…
Tensorlake introduced dynamic network policies for its Sandboxes product, enabling AI agent workloads to change network boundaries during execution phases—trusted context, approved tool, network isola…
Tensorlake's Cloud Volumes enable a reproducible five-agent pipeline by versioning the filesystem instead of teaching agents Git, allowing every run to be snapshotted, diffed, and restored on a fresh …
Tensorlake's snapshot-and-fork feature lets users clone a fully provisioned AI agent sandbox into multiple workers, cutting setup time from 12.3 seconds per worker to a one-time 14.0 seconds plus 1.78…
Tensorlake has introduced Sandboxes, an isolated execution environment for AI workloads, enabling developers to run Python code, manage files and dependencies, and preserve runtime state for productio…
Tensorlake's Git Repositories feature allows AI agents to mount versioned repos as local directories without cloning, using a FUSE filesystem and write-ahead logging for crash safety. In tests with CL…
OpenCode, an open-source coding agent similar to Claude Code, poses a risk because it autonomously runs shell commands, which can inadvertently execute destructive actions like deleting files via syml…
Tensorlake's memory snapshots let AI experiments resume from a warm state, eliminating the repeated 40-second overhead of reinstalling numpy and reloading datasets that occurs when using filesystem sn…
A new open-source harness from Tensorlake demonstrates that agent evaluations are vulnerable to cheating unless per-task isolation is enforced, showing a 53% lie rate when agents can tamper with test …
A developer built a stateful AI agent that survives session kills by using Tensorlake's MicroVM sandboxes instead of containers. The harness preserves running processes, files, and memory across suspe…
A developer built a five-agent data analysis pipeline on Tensorlake and confirmed that the platform's MicroVM-based sandbox isolation prevents agent crashes, filesystem conflicts, and memory contentio…
A developer lost six hours of agent work after assuming Claude Managed Agents would retain state between sessions, discovering that Anthropic's infrastructure handles reasoning while users must manage…
Engineers often assume multi-agent AI performance is a concurrency problem, but the real bottleneck is agents spending most of their time preparing to work rather than working. The fix is using memory…