Priced per robot, not per gigabyte.

Telemetry pricing punishes you for capturing the thing that explains the failure. Watchpoint charges by fleet size, so the incentive to instrument deeply stays intact.

Community

Free
self-hosted, forever

The full forensics core, running entirely on your infrastructure.

  • Unlimited devices and users
  • Model collector with PyTorch adapter
  • ROS 2 collector and Go edge agent
  • 7 system rules + shipped AI rules
  • Incident timeline and inference views
  • Replay bundle ZIP export
  • Community support via GitHub issues
Read the source

Team

Most robotics teams
$49
per robot / month, billed annually

For fleets past the point where one engineer can triage by hand.

  • Everything in Community
  • Full AI rule taxonomy as it ships
  • Deterministic replay sandbox (roadmap)
  • Attention overlays and decision traces (roadmap)
  • Cross-fleet baselines and regression grouping
  • Guided upgrades and migration support
  • Priority support, 1 business day response

Minimum 10 robots. Robots in a bench or CI fleet aren't counted.

Talk to us

Enterprise

Custom
annual contract

For teams with a safety case, an auditor, or an air gap.

  • Everything in Team
  • SSO / SAML and audit logging
  • Air-gapped and on-prem deployment support
  • Incident record exports for safety cases
  • Custom rules built for your failure modes
  • SLA with named support engineer
  • Roadmap input and design review
Contact sales
Design partner program — open now

Free for 12 months. We build your failure mode.

We're taking a small number of design partners running real fleets. You get the Team tier at no cost for a year, and we implement the failure mode that costs your team the most time as a first-class rule in the engine.

In exchange we ask for honest feedback, an hour every two weeks, and permission to reference your team once you're happy with the results.

Good fit if you
  • Run 5+ robots on ROS 2 or a custom autonomy stack
  • Have learned perception in production today
  • Lost more than a day to a field incident recently
  • Can self-host a Postgres and a container
Apply as a design partner
Hosted cloud — not yet available

We deliberately built self-hosted first, because most robotics teams cannot export customer camera data. A managed option is planned for teams without that constraint. It isn't built yet, and we'd rather say so than take a waitlist deposit on it.

Tell us you want it

What's in each plan

CapabilityCommunityTeamEnterprise
Self-hosted deployment
Model collector + ROS 2 + edge agent
System rules engine (7 rules)
Shipped AI rules (AI-001–003)
Replay bundle export
Full AI taxonomy (AI-004–008)roadmap
Replay sandboxroadmap
Cross-fleet baselines
Priority support SLA
SSO / SAML + audit log
Air-gapped deployment support
Custom rules built for you

Items marked roadmap are specified but not merged. Team and Enterprise include them as they ship; we don't bill for them before that.

Questions

What exactly counts as a robot?

A physical unit running a collector and reporting under its own device ID. Bench rigs, simulation, and CI fleets are free and uncounted — we don't want to price you out of testing.

Is the Community tier crippled?

No. It's the complete forensics core: every collector, the system rules engine, the AI rules that have shipped, and bundle export. Paid tiers add fleet-scale analysis, the replay sandbox, and support — not access to your own data.

What's the license?

The core is Apache 2.0 — every collector, the API, the rules engine, and the dashboard. Anything that runs on your robot is Apache-licensed without exception, so you can audit it before you deploy it. Commercial features like SSO and cross-fleet rollups are licensed separately and live outside that tree.

Do you ever see our camera data?

No. Self-hosted means the stack runs in your VPC with your Postgres and your storage. Model weights are hashed for lineage, never uploaded. With no Anthropic key set, the LLM summary degrades to deterministic rules text and nothing leaves your network at all.

What happens if we stop paying?

You fall back to Community. It's self-hosted, so your data stays where it already is — there is no hostage situation and no export scramble.

Why is pricing per robot instead of per data volume?

Volume pricing makes you capture less exactly when you should capture more. The whole product depends on the failure frame being in the buffer, so we refuse to charge in a way that discourages that.