HomeBlogPricingCareersDocsGitHubSlack community
Talk to FounderDashboard →
Field notes/Community builds/Snapshot Once, Rollout a Thousand Times: A Practical RL Setup for Coding Agents

Snapshot Once, Rollout a Thousand Times: A Practical RL Setup for Coding Agents

Snapshot the environment once and fork every rollout from it: 8,000 RL rollouts drop from ~20 hours of naive environment rebuilds to ~4.4 hours.

SBSebastian Buzdugan·JUN 17, 2026·11 MIN READ·Community builds
SBX-01C4SBX-01E3SBX-0202SBX-0221SBX-0240SBX-025FSBX-027ESBX-029DSBX-02BCSBX-02DBSBX-02FASBX-0319SBX-0338SBX-0357SBX-0376[ RUNTIME: ACTIVE ] P50 2.45S · P99 4.12S · 5M/PROJECT

The RL-training version of the snapshot argument: build the environment once, snapshot it, and fork every rollout from that image. Across 8,000 rollouts for a coding agent, Sebastian's setup drops environment time from roughly 20 hours of naive rebuilds to about 4.4 hours — the difference between an overnight run and a same-day iteration loop.

We didn’t write this one — it’s Sebastian Buzdugan’s piece, published on Data Science Collective. The note above is ours; the full article is theirs.

Read the full piece on Data Science Collective ↗
SB
WRITTEN BYSebastian BuzduganCommunity · Data Science Collective
Read next —FROM THE LOG
Jun 24, 2026Engineering

Accelerating Harbor with Tensorlake.Tensorlake runs the real task Docker images for Harbor on microVM sandboxes. On Terminal-Bench 2.1 the oracle passes all 89 tasks in 36m 44s — every image published, cold starts in seconds.

Shanshan WangApplied AI · Tensorlake6 minJun 5, 2026Engineering

Tensorlake Sandboxes as the runtime for Claude Managed Agents.Run Claude Managed Agents on Tensorlake Sandboxes: you keep the control of self-hosting, and drop the always-on box. The short version — the server goes away.

Shanshan WangApplied AI · Tensorlake7 minJul 9, 2026Engineering

Zero-copy TLS ingress with kTLS and splice(2) for sandboxes

Diptanu Gon ChoudhuryCEO & Co-founder8 minJul 13, 2026Engineering

Firecracker disk snapshots in O(changed bytes), not O(disk size)We added copy-on-write image layering inside Firecracker's block device, and the disk got faster, not slower. Snapshots pause the VM for tens of milliseconds and cost only what changed, so a running Postgres barely notices them, at any disk size.

Diptanu Gon ChoudhuryCEO / Co-founder10 min
◆ THE SANDBOX DIGEST

Subscribe for release notes, benchmarks, deep dives.

One dispatch per month from the Tensorlake team — no spam.