Snapshot
Continue to About ↓- What it does
- boat rents persistent Linux (Ubuntu) cloud virtual machines called sandboxes, built for running AI coding agents and for human developers.
More detail
- Each sandbox is a full VM with SSH/SCP, Docker, a virtual desktop and snapshot-based forking, driven by CLI, HTTP API or Python/TypeScript SDKs and billed per second while running.
- Best for
- Teams operating many parallel AI coding agents or running untrusted code, builds, tests and Docker workloads in isolated cloud VMs.
More detail
- Cost-sensitive users who want full Ubuntu machines billed per second rather than fixed hourly commitments.
- Pricing
- 7-day free trial with 25 compute hours, then a $20/month minimum plan that buys 2,000,000 VM-seconds (about 555 hours) of default 4 vCPU / 8 GB time, billed per second only while a sandbox runs.
More detail
- Sizes scale the burn rate: small (2 vCPU / 4 GB) at 0.5x, default at 1x, large (8 vCPU / 16 GB) at 2x; li
- Main limitation
- Buyers who need to keep data in the US or outside the EU, since sandboxes run only in Germany, Finland and France today.
More detail
- Low-usage users who want to spend under $20 a month, or anyone needing a permanent free tier or more than 2,000 self-serve concurrent sandboxes.
- Record
- Living and monitored
- Reviewed
- Evidence
- Public web research
- Machine access
- Structured profile ↗
About
- boat rents persistent Linux (Ubuntu) cloud virtual machines called sandboxes, built for running AI coding agents and for human developers.
- Each sandbox is a full VM with SSH/SCP, Docker, a virtual desktop and snapshot-based forking, driven by CLI, HTTP API or Python/TypeScript SDKs and billed per second while running.
What it helps with
- Spinning up and managing many isolated cloud machines for AI agents without building the infrastructure yourself.
- Needing a full Ubuntu machine (Docker, systemd, databases, cron) rather than a lightweight container sandbox to run any
- Keeping untrusted code, builds, tests and long jobs off local developer machines.
- Giving agents machine-readable docs and JSON output so they can provision and control sandboxes themselves.
- Cloning environments quickly from snapshots to run parallel agent work.
- Controlling cost on sporadic agent workloads by paying only while a sandbox runs.
Capabilities
A practical view of what is available and where limitations apply.
| Offering | Available | Detail |
|---|---|---|
| Full Ubuntu VM sandboxes | Yes | Each sandbox is a persistent Linux VM with SSH/SCP, Docker support, a dedicated IPv4/IPv6 address and a 60fps virtual desktop. |
| Pre-installed developer tooling | Yes | Docker, VS Code, Chrome, Ghostty, GitHub CLI, Rust, Node.js and Bun come pre-installed per the official homepage. |
| Snapshot forking | Yes | Disk-level forking from a snapshot; saved snapshots are listed at $0.056 per day each. |
| Programmatic access | Yes | HTTP API, Python and TypeScript SDKs, plus a CLI that accepts --json for JSONL output. |
| Agent integrations | Limited | Docs list Claude Code, Codex and Cursor (via MCP server); the set of supported agents is not stated as exhaustive. |
| US or non-EU hosting regions | No | Official site states EU regions today only: Germany, Finland and France. |
| Free plan | No | A 7-day trial with 25 compute hours is offered, but plans start at a $20/month minimum. |
| Independent performance verification | Unknown | Homepage speed comparisons against Daytona, E2B, Modal and others are first-party claims; no independent replication was found in the evidence. |
| Independent customer reviews | Unknown | Only a quoted homepage testimonial was found; no independent review source appears in the evidence. |
| Scale beyond 2,000 concurrent sandboxes | Limited | Self-serve plans run 100 to 2,000 active sandboxes per account; higher counts require a custom agreement. |
Bot trust
How confidently AI systems can verify this profile from public evidence.
54of 100
ChatGPT—/100
Claude54/100
What bots could verify
- • Corporate identity is clearly documented, identifying parent entity Dedale AI, Corp., leadership, and location.
- • Pricing model, included VM-seconds, tier multipliers, free trial limits, and refund policies are explicitly detailed.
- • External technical evaluation is cited via Legit.Show benchmarking performance, accessibility, and web standards.
What weakens trust
- • Almost no independent customer feedback, case studies, or verified platform reviews (e.g., G2, Trustpilot, community forums).
- • Security certification is incomplete with SOC 2 only in progress, and third-party scans found missing security headers (CSP, HSTS).
- • Extremely brief operational history and small team size create substantial reliance on an unproven startup track record.
Evidence
Where this record came from, when it was reviewed, and how confidently it can be used.
- Verification
- Scribble researched these facts from the linked public sources.
- The owner has not manually confirmed every field.
- Last reviewed
- Primary source
- boat.dev
Technical profile details
Record ID
5d5f7d0d-5878-4e47-b7ce-beb0290e063eCanonical URLhttps://profiles.scribble.network/listing/boat-5d5f7d