HomeBlogPricingCareersDocsGitHubSlack community
THE INFRASTRUCTURE LAYER FOR THE AGENTIC STACK

Composable infrastructure primitives for agents.

Isolated sandboxes, hosted Git, and a versioned filesystem for deploying agent harnesses, running RL rollouts and evals, and building CI that scales.

Free tier included. No credit card required.

● RUNNING
import { Sandbox } from "tensorlake";

async function main() {
const sbx = await Sandbox.create();
const result = await sbx.run("/bin/sh", { args: ["-c", "npm install && npm run build"] });
console.log(result.stdout);
}
main();
$npx tsx sandbox.tsRAN
✓ sandbox created sbx_rhcj28c · tensorlake/ubuntu-minimal · 460 ms
✓ npm install (143 pkgs) · 1.7 s
✓ npm run build · compiled in 4.8 s
SandboxesIsolated microVMs
SnapshotsPause, fork and resume in seconds
Cloud VolumesVersioned, mountable shared directories
Serverless FunctionsDistributed, workflows and durable
Cloud Dev BoxesSSH, IDEs and Persistence
EXECUTION ENVIRONMENTS02 / 07

Give your agent a computer.

PATTERNS
harness · tool

Run persistent sandboxes that give an agent its own computer, or spin up ephemeral ones to isolate each tool call.

ISOLATED RUNTIME

Give the agent its own filesystem, shell, packages, and processes instead of sharing the app server runtime.

STATEFUL BY DEFAULT

Sandboxes sleep on inactivity and wake instantly when invoked — sessions can run for hours or days.

REAL SOFTWARE STACKS

Compile code, run databases, process large files. Bring any Linux stack.

RUN TOOL CALLS SAFELY

Execute LLM-generated code inside secure microVMs, isolated from the harness.

POSIX directory mounts and hosted Git so your agents never lose the work they produce.

tl fs · MOUNT · AUTOSAVE · SNAPSHOT

A versioned filesystem you mount like a disk.

Mount a durable volume and it shows up as an ordinary POSIX directory. Reads and writes hit local NVMe, and every settled change saves itself to durable storage. Take a snapshot to pin a permanent checkpoint, roll back to any point in time, and pick the work back up on another machine after a crash.

MOUNT ANYWHERE

A POSIX directory on local NVMe. It runs on any sandbox, cloud VM, or Kubernetes pod, on Linux or macOS.

SAVES ITSELF

Settled writes replicate to durable storage on their own, so an agent keeps a versioned working state without ever learning Git.

SNAPSHOT AND TIME TRAVEL

Seal a permanent checkpoint with one command. Restore or fork the whole tree from any point you kept.

SHARE READ-ONLY

Hand skills, tools, and configs to a whole fleet with read-only mounts that follow head or pin a snapshot.

tl fs · SESSION● agent-scratch
$ tl fs mount agent-scratch ~/work✓ mounted · session 54398548 · autosave on (~1s)# a normal POSIX directory. writes hit local NVMe$ echo "parser is quadratic" > ~/work/notes.md↻ autosave · settled changes replicate to durable storage$ tl fs snapshot ~/work -m "baseline"✓ snapshot 4d9a2f7e · permanent checkpoint$ tl fs restore agent-scratch @2m-ago✓ time-traveled · mount now at an earlier state↺ crash the sandbox, remount anywhere, and nothing is lost
tl git · CLONE · SNAPSHOT · PROMOTE

Managed Git that scales to millions of repos.

Use it as ordinary Git and clone, branch, commit, merge, and push the way you always have. Or mount a repository into a sandbox with no clone at all: work in a private workspace, take snapshots as you go, and promote the result to a branch with a compare-and-swap. Merges run on the server, and you pay for the paths that changed rather than the size of the repo.

PLAIN GIT

Clone, branch, commit, merge, and push over Git smart HTTP. It works with the tools your team already uses.

MOUNT WITHOUT CLONING

Stream a branch into a sandbox on demand. Content arrives as it's read, so there's no full checkout to wait on.

SNAPSHOT THEN PROMOTE

Autosave protects a private workspace WAL. A snapshot seals a commit, and promote moves a branch with a compare-and-swap that copies zero bytes.

SERVER-SIDE MERGES

Merge, rebase, and preflight conflicts without a checkout, at hundreds of pushes per second per repository.

tl git · WORKSPACE● ws_3f9a2b7e
$ tl git mount agent-outputs ./work✓ mounted · workspace ws_3f9a2b7e · no clone, streams on read# build tools run on local NVMe$ npm install && npm run build✓ 143 pkgs · build ok$ tl git snapshot ./work -m "fix parser"✓ snapshot s3 · sealed as a Git commit$ tl git promote ./work main✓ main → s3 · CAS ok · 0 bytes copied↳ git fetch sees it. it's a normal commit
THE GIT MODEL · WORKSPACE = REF · SNAPSHOT = COMMITbranch pointers move; data never does
mainbranchworkspacerefs/workspaces/ws_7fq2PROMOTE · CASC1C2 · baseC3 · squashs1s2s3 · snapshot

A workspace is just a ref, and each snapshot is an ordinary commit on it. Promote moves main with a compare-and-swap, so no bytes are copied and no trees rebuild. If main has moved, the promote either fails or merges, and the snapshot that lost stays a recoverable commit you can reach through workspace and operation history.

[ TWO SURFACES · ONE STORAGE BACKEND ]

Store anything an agent produces. Version all of it.

01GIT

Generated code

One repository per generated project, with branches, merges, and a durable history that records who changed what. Keep it somewhere you can publish from.

02FS

Generated artifacts

Model outputs, reports, and datasets stay versioned and easy to roll back, so they don't disappear when the sandbox goes away.

03FS

Documents

Long-lived working documents with full history, restore, and undo across every editing session.

04FS

Session memory

Mount ~/.claude and snapshot after each phase. Agents resume exactly where they left off, on any machine.

SNAPSHOT · RESUME · COPY03 / 07

Suspend, resume, and copy a whole machine.

LIFECYCLE
suspend · resume · fork
SUSPEND · RESUME

Pause on idle. Wake in a second.

Freeze a sandbox — memory, processes, and filesystem intact — and the meter stops. Resume by name in about a second, or let it wake automatically the moment a request or connection comes in.

LIFECYCLE · CLItl sbx
$ tl sbx suspend my-dev
✓ suspended · meter stopped · state on disk
$ tl sbx resume my-dev
✓ resumed in 0.9s · pid 4127 still alive
↺ auto-wakes on the next request or connection
COPY · FORK THE VM

Fork a running VM, memory and all.

Clone a live sandbox — its memory and filesystem copied whole — then run the copies. A fork of the entire machine: build a base once, then fan out identical copies to run many CI jobs in parallel.

COPY · CLItl sbx copy
$ tl sbx copy build-base --count 8
✓ forked 8 copies (memory + filesystem)
build-base-01 … build-base-08 running
$ tl sbx ls --tag build-base
8 sandboxes · 0 cold starts · ready for jobs
BENCHMARKS04 / 07

The fastest sandbox file system.

METHOD
fio · sqlite · p50

SQLite benchmark — 2 vCPU, 4 GB RAM, 100k inserts

View benchmark on GitHub
Tensorlake
2.45s1.0×
Vercel
3.00s1.2×
E2B
3.92s1.6×
Modal
4.66s1.9×
Daytona
5.51s2.2×
FSYNC4.1×
SEQ WRITE2.8×
RAND READ1.9×
COLD START84ms

In our published SQLite benchmark across Tensorlake, Vercel, E2B, Daytona, and Modal, Tensorlake was the fastest across default, fsync, and large-dataset runs. Benchmark setup: 2 vCPU / ~4 GB sandboxes, 3 runs.

HARBOR · EVALS & RL05 / 07

Harbor integration for agents evals and RL rollouts.

BENCH
terminal-bench@2.1
01

Run your Docker image, unchanged

Import any OCI image and convert it to an ext4 rootfs with full fidelity — filesystem and runtime metadata preserved. Sandboxes boot exactly like your Docker container, no translation errors.

02

Pre-published, boots in seconds

Images are already there — no per-run build, no first-run pull-and-cache. A multi-minute build becomes a seconds-long boot. That fidelity plus speed is what took the oracle to 1.000.

03

Snapshot & scale rollouts

Snapshot a warmed environment, then clone and fan out to 10k+ concurrent sandboxes. Checkpoint and resume at any step, write to object storage.

HARBOR · RUNterminal-bench@2.1
# pre-published image — boots directly, no rebuild
harbor run --env tensorlake \
  --dataset terminal-bench/terminal-bench-2-1 \
  --agent claude-code \
  --model anthropic/claude-opus-4-8 \
  --ae ANTHROPIC_API_KEY=$ANTHROPIC_API_KEY

# oracle pass — reproduces the run on the right
harbor run --env tensorlake \
  --dataset terminal-bench/terminal-bench-2-1 \
  --agent oracle
OUTPUT● 1.000
terminal-bench • oracle
┏━━━━━━━━┳━━━━━━━━━━━━┳━━━━━━━┓
┃ Trials ┃ Exceptions ┃  Mean ┃
┡━━━━━━━━╇━━━━━━━━━━━━╇━━━━━━━┩
│     89 │          0 │ 1.000 │
└────────┴────────────┴───────┘

┏━━━━━━━━┳━━━━━━━┓
┃ Reward ┃ Count ┃
┡━━━━━━━━╇━━━━━━━┩
│ 1.0    │    89 │
└────────┴───────┘

Job Info
Total runtime: 36m 44s

Tensorlake 1.000 · 36m 44s — 1.7× faster than the next best provider.

ORCHESTRATION06 / 07

Sandbox-native orchestration for agents.

LAYER
endpoints · durability
[ 06.1 ] · ORCHESTRATE

Once sandbox usage turns into a real application, Orchestrate coordinates it.

The layer that adds application endpoints, durability, fan-out, retries, and application-level observability on top of sandbox execution.

APPLICATION ENDPOINTS

Expose sandbox-backed workflows as callable applications instead of stitching together raw VM APIs.

DISTRIBUTED FAN-OUT

Predictable throughput means fresh sessions spin up immediately, even when a thousand others are mid-task.

WAKE ON REQUEST

Dormant sandboxes resume on incoming traffic. Every session gets its own sandbox so nothing leaks across runs.

QUEUES, TIMERS, AND RETRIES

Durable primitives for long-running agentic flows. Application observability baked in.

ORCHESTRATE · APP | doc-to-md$ tl app deploy
# PDF → Markdown with Claude
from tensorlake.applications import application, function
from anthropic import Anthropic

claude = Anthropic()

@application()
@function()
def to_markdown(pdf_url: str) -> str:
pdf = fetch(pdf_url)
msg = claude.messages.create(
model="claude-sonnet-4-5",
max_tokens=8192,
messages=[{"role": "user", "content": [
{"type": "document", "source": pdf},
{"type": "text", "text": "Convert to clean Markdown."},
]}],
)
return msg.content[0].text
REMOTE DEV07 / 07

SSH in from any machine. The sandbox is your dev environment too.

ACCESS
ssh · vs code · pty

A named sandbox isn't only an execution surface — it's a portable cloud workstation. Register your key once, then ssh in, or open it in VS Code, Cursor, or JetBrains Gateway over Remote-SSH. scp, rsync, and full port-forwarding all ride the same connection.

PORTABLE IDENTITY
Connect to the same sandbox id from any laptop. Which region or host the VM lives on is the platform's problem — not yours.
SUSPEND ON IDLE
Named sandboxes auto-suspend after the idle timeout. The clock pauses while you're connected, so you only pay while you work.
STATE SURVIVES RESUME
Filesystem, memory, running processes, tmux sessions and ~/.vscode-server are exactly where you left them — resume by name, weeks later.
PTY OVER WEBSOCKET
Building a browser terminal? Drive a programmatic PTY over WebSocket, disconnect, and reattach to the same session by token.
REMOTE DEV · WALKTHROUGH● 0:45
LOCAL — ZSH
~/PROJECTS/TENSORLAKE
tensorlake
BACKED BY
DEPLOY & TRUST08

Run it in our cloud — or yours.

COMPLIANCE
SOC 2 · HIPAA
[ 08 ] FOR TEAMS OUTGROWING SAAS

Bring Tensorlake into your cloud.

Run sandboxes and applications inside your own AWS / GCP / Azure account when you need lower egress, stricter network boundaries, dedicated capacity or more predictable performance.

01
Network boundariesKeep code and data inside your preferred cloud boundary. VPC peering, private endpoints, IAM-scoped.
02
Latency controlCompute closer to data. Tighten runtime behavior for latency-sensitive agent workloads.
03
Reserved capacityMove from usage-based hosted infra to capacity you plan, reserve, and operate predictably.
[ 09 ] SECURITY

Security built for agentic workflows.

LLM-generated code runs in isolated VMs, not shared processes. Full audit trails, per-project data boundaries, and compliance for regulated workloads.

01
Firecracker isolationEach tool call and harness runs in its own microVM. No shared kernel, no cross-tenant state.
02
Tracing & observabilityFull traces of every function and tool call — logs, timing, structured execution paths.
03
SOC 2 Type II · HIPAASecure by default for PHI, PII and sensitive documents. Isolated buckets with RBAC and full audit.
▪ THE DOCUMENT DIGEST · WEEKLY

Subscribe for release notes, benchmarks, deep dives.

~ 3,400 ENGINEERS. MONTHLY. NO SPAM.
▪ THE INFRASTRUCTURE LAYER FOR AGENTS

Ship agents faster
with Tensorlake.

$pip install tensorlake #  or:  npm i @tensorlake/sdk