The work begins with a blank model. A systems engineer opens a new project in Clear, Safe Robots’ product, and starts by dragging in the robot’s sensor list from a GitHub repo. They link a regulatory document from Drive, a set of functional requirements from Jama, and a few open Jira tickets. The tool doesn’t just store these artifacts. It begins to draw lines between them, constructing a map of what must be proven and what evidence exists to prove it. For a team trying to ship a physical machine that won’t hurt anyone, this map is the only thing standing between a prototype and a product.
The Wedge Is the Paperwork
Safe Robots is not building the robots. It is building the safety case for them, a dense thicket of documentation that is both a legal necessity and a traditional bottleneck. The company’s bet is that this paperwork, often a manual afterthought, can be transformed into a dynamic, connected system,a “living model.” The product, Clear, acts as a central nervous system for safety engineering, plugging into the tools teams already use for code (GitHub), project management (Jira), and requirements (Jama) to maintain traceability automatically [F6S, September 2025]. The promise is to let robotics teams “deploy fast” without cutting corners on the rigor demanded by insurers, customers, and regulators in automotive, medical, and industrial robotics [Safe Robots].
A Founder Shaped by Cruise
The company’s technical vision is deeply informed by the lived experience of its co-founder and CEO, Nestor Grace. For nearly three years, Grace worked as a Senior System Engineer at Cruise, the GM-backed autonomous vehicle company. His public profile outlines a role immersed in the gritty reality of safety-critical deployment: requirements management, safety analysis, testing, and failure-mode monitoring for robotaxis [LinkedIn, October 2026]. This is the arena where theoretical safety meets operational reality, where a missing link in a requirements traceability matrix can stall a multi-million dollar program. Grace’s co-founder, Thejas Kesari, is described as a product builder with a decade of experience in safety-critical software [F6S, September 2025]. The team’s pivot to focus exclusively on robotics and autonomy came in July 2025, a few months after Grace’s tenure at Cruise ended [Anthony B - Stealth | LinkedIn, October 2025].
How Clear Attempts to Automate Rigor
The product surfaces a series of interconnected modules designed to methodically build the safety argument. It’s a process that moves from understanding the system to proving it’s safe.
System Modeling | 1 | Core Module
Regulatory Analysis | 1 | Core Module
Scenario Mapping | 1 | Core Module
Hazard & Risk Analysis | 1 | Core Module
Compliance & Certification | 1 | Core Module
- System Modeling. This creates a single source of truth for what the robot senses, does, and operates within, forming the foundation for all subsequent analysis [Safe Robots].
- Regulatory Analysis. The tool identifies the specific rules and standards that apply based on where and how the robot will be deployed, converting often-opaque regulations into actionable items [Safe Robots].
- Scenario Mapping. Instead of relying on engineers to brainstorm every possible situation, Clear generates potential scenarios a robot could encounter and measures test coverage against them [Safe Robots].
- Hazard & Risk Analysis. It helps teams identify potential failures, assign them severity ratings, designate owners, and roll up risk across the entire system into an auditable register [Safe Robots].
- Compliance & Certification. This module aims to thread the needle from binding standards to testable requirements, maintaining the continuous traceability needed for certification audits [Safe Robots].
The through-line is the living model. A change in a source code file flags which requirements are now unverified; a new hazard logged in Jira automatically updates the risk register. The intent is to make safety engineering a continuous, integrated discipline rather than a monolithic, pre-launch burden.
The Early-Stage Landscape
Operating out of San Francisco and backed by the Berkeley SkyDeck accelerator, Safe Robots is navigating a market that is recognizing the scale of the problem [F6S, September 2025]. The space for robot safety tooling is beginning to attract attention and capital. Notably, a separate startup named SafeWorld emerged from stealth in October 2026 with $12.2 million in seed funding to build simulation software for stress-testing robots [The AI Insider, October 2026]. While both target robot safety, their approaches appear divergent: SafeWorld focuses on simulation-based testing, while Safe Robots is focused on the engineering lifecycle and documentation rigor. This suggests a market bifurcating between proving safety in simulation and managing the evidence for safety in reality.
| Company | Primary Focus | Key Differentiator | Recent Funding |
|---|---|---|---|
| Safe Robots | Safety & Quality Engineering Lifecycle | “Living model” for requirements, hazard analysis, and traceability | Berkeley SkyDeck accelerator backing [F6S, September 2025] |
| SafeWorld | Simulation-Based Testing | Digital humans and environments to stress-test robot behavior | $12.2M Seed Round [FinSMEs, October 2026] |
Where the Friction Will Be
The ambition is clear, but the path for a pre-seed startup is lined with formidable challenges. The first is credibility. Selling into safety-critical industries requires an immense depth of domain expertise and a product that is, itself, certifiably robust. Early adopters will be taking a risk on a new tool for their most risk-averse process. Second, the integration-heavy model means Clear’s utility is directly tied to its ability to seamlessly connect with and correctly interpret data from a widening array of engineering tools, each with its own quirks and update cycles. Finally, the team must navigate a likely name collision with SafeWorld and a separate UK entity, ensuring their market message cuts through the noise [Perplexity Sonar Pro Brief; Companies House, September 2025].
The company’s answer to these hurdles appears to be founder-led domain expertise and a sharp focus on the workflow itself. Grace’s background is a direct response to the credibility question. The focus on integrations like GitHub, Jira, and Jama is a pragmatic embrace of the existing environment, not a demand to overhaul it [F6S, September 2025]. The bet is that engineers who have lived through the pain will recognize the tool built by someone who has, too.
The Next Twelve Months
The immediate milestones are classic for a company at this stage: convert pilot users into paying customers, expand the integration ecosystem, and likely secure a substantive seed round to scale the team. The validation will come not from a flashy headline number, but from a quieter signal: a robotics or automotive supplier signing a purchase order because Clear demonstrably shaves weeks off their certification timeline. The goal for Safe Robots is to become the unspoken prerequisite, the layer of necessary rigor that ambitious hardware teams adopt not as a luxury, but as the only sane way to build.
It starts with a blank model, but the question it ultimately answers is cultural. For decades, speed and safety in hardware have sat on opposite sides of a trade-off. You could move fast, or you could be sure. What Safe Robots is testing is whether that trade-off is a law of physics or just a limitation of outdated tools. The living model is an argument that the paperwork isn’t the enemy of progress, but its only reliable blueprint. The company’s success hinges on convincing a generation of engineers that the most important thing their robot will ever build is the case for its own existence.
Sources
- [F6S, September 2025] Safe Robots company profile | https://www.f6s.com/company/try-clear?flow=seePage
- [LinkedIn, October 2026] Nestor Grace profile | https://www.linkedin.com/in/nestor-grace
- [Safe Robots] Company homepage | https://www.saferobots.co/
- [Anthony B - Stealth | LinkedIn, October 2025] Post on Safe Robots pivot and Demo Day | https://www.linkedin.com/in/anthony-bo/
- [The AI Insider, October 2026] SafeWorld Emerges From Stealth With $12.2M in Funding | https://theaiinsider.tech/2026/10/07/safeworld-emerges-from-stealth-with-12-2m-in-funding-to-build-robot-safety-simulation-software/
- [FinSMEs, October 2026] SafeWorld Raises $12.2M in Seed Funding | https://www.finsmes.com/2026/10/safeworld-raises-12-2m-in-seed-funding.html
- [Perplexity Sonar Pro Brief] Analysis on name collision | (source analysis)
- [Companies House, September 2025] SAFE ROBOTICS UK LIMITED overview | https://find-and-update.company-information.service.gov.uk/company/16727334