Velaura AI
Functional Safety Architect
$200K–$300K
Santa Clara, CA · On-site · Full-time · Posted Jul 27, 2026
- Experience
- 10+ years in functional safety or safety-critical hardware engineering, with significant experience taking silicon or embedded platforms through a full safety lifecycle.
- Education
- Bachelor’s or advanced degree in Electrical/Computer Engineering, Computer Science, or a related field, or equivalent practical experience.
About the job
Own the functional safety architecture and lifecycle for Velaura’s Physical AI SoCs, from early design through certification and production.
What you'll do
- Define SoC safety goals, mechanisms, fault handling, and quantitative safety metrics; lead safety analyses.
- Integrate safety requirements across silicon, firmware, and platform software, and own the safety case and lifecycle work products.
- Establish safety processes, coordinate certification and audits, and mentor engineers while managing safety interfaces with cybersecurity, reliability, and SOTIF.
What they ask for
- Deep expertise applying ISO 26262 up to ASIL D and IEC 61508 up to SIL 3 to SoC development.
- Hands-on leadership of safety analyses and hardware architectural metrics, including SPFM, LFM, and PMHF.
- Strong command of hardware safety mechanisms and the interface between RTL/silicon, firmware, and low-level platform software.
- Proven ability to build and defend a safety case and drive cross-disciplinary safety requirements from architecture through production.
Nice to have
- Experience with safety for AI/ML accelerators or NPUs and SOTIF (ISO 21448).
- Familiarity with robotics, automotive, industrial, or other safety-critical domains; adjacent standards; or certification bodies and assessors.
Pay
USD 200,000–300,000 per-year-salary (base salary)
Tools
ISO 26262, IEC 61508, RTL, ECC, lockstep, BIST, NPU, ISO 21448
AI focus
Define and assure functional safety for AI/NPU compute in Physical AI SoCs, including safety mechanisms for AI/NPU subsystems.
Model-written summary · the original posting is authoritative