SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Salary: USD 182,000 - 230,000 / annual
Atoms is building Physical AI—real-world robots for industries including food, mining, and transport. The company integrates hardware, software, AI, operations, and manufacturing to deploy autonomous systems at scale.
As Safety Case Engineer, you will own a defined scope of the safety case end-to-end, from initial drafting through sign-off and continuous governance as upstream dependencies evolve. You will serve as author of record, holding ultimate accountability for defending the safety argument.
Key responsibilities include:
• Architect clear argument boundaries and constraints, defining operational conditions, valid assumptions, explicit exclusions, and non-credible hazards within your scope.
• Evaluate evidence sufficiency and coverage, scrutinizing operating experience metrics and data provenance while partnering with Functional Safety, Scenario Validation, and Fleet Operations leads to synthesize required evidence.
• Construct the Claim-Argument-Evidence (CAE) structure, establishing precise validity boundaries so it is transparent exactly when safety claims hold and when they do not.
• Lead formal safety reviews and peer governance, convening reviews only when evidence meets maturity standards, driving reviewer feedback to documented resolution, and serving as an independent reviewer for peer scopes.
• Drive rigor in shared methodologies and tooling, proactively identifying process gaps or ambiguous rules to prevent local workarounds from compromising organizational safety standards.
• Pioneer AI-assisted safety engineering workflows, leveraging AI tools to draft from templates, stress-test candidate hazard sets, cross-check scenario coverage, maintain end-to-end traceability, and rapidly flag invalidated claims.
The role is based in San Francisco and requires onsite presence five days per week.
REQUIREMENTS:
• 9+ years of cross-functional experience spanning functional safety, autonomous behavior, and fleet operations, with deep expertise in at least two areas and working fluency in the third.
• Proven track record as author of record for a structured safety argument (GSN or CAE equivalent) on a production system, including defending claims through rigorous technical scrutiny.
• Dual ownership of hazard analysis and validation, with demonstrated mastery in safety analysis methods (HARA, FMEA, FTA) paired directly with scenario-based validation strategies.
• Uncompromising rigor regarding data provenance, with an instinctive drive to distinguish, label, and challenge measurements, estimates, and placeholders before they acquire false authority.
• Courage and integrity to enforce safety gates, comfortably speaking up when a safety case is unready—even under commercial pressure—while clearly articulating missing evidence and associated risks.
• Proportional mastery of safety standards, applying ISO 26262, ISO 21448 (SOTIF), UL 4600, and SAE J3016 appropriately to scoped arguments without imposing unnecessary process overhead.
• Statistical fluency in exposure, rare events, and risk metrics, including a clear understanding of surrogate safety measures and their technical limitations when evaluating harm.
• Functional programming proficiency (e.g., Python) to analyze fleet data, engineer local tooling, and leverage AI assistants effectively—maintaining firm boundaries where AI-generated outputs are treated strictly as hypotheses or drafts.
• Exceptional written precision, producing clear, unambiguous safety documents capable of standing up under independent scrutiny without author intervention.
• Collaborative leadership and team flexibility, balancing deep ownership of assigned scope with agility to support peer scopes and provide high-value, non-ceremonial reviews.
