About the Role
Uber is launching AV Labs to accelerate the autonomous technology ecosystem. We're building out a high-velocity team of multi-disciplinary experts to turn real-world operations into high-quality data for our autonomous partners. This team is focused on the hardest problem in AV today: unlocking real-world, long-tail driving data. Autonomy is now a data race—and Uber has an edge: We collect rare, real-world driving data at a scale and capital efficiency no one else can match.
Our AV Labs Hardware team is building the next-generation E/E architecture for our L4 data-collection vehicle fleet. A fleet only collects useful data when its compute, MCU and sensors work reliably in the car, and when a fault can be found and fixed without a technician on site. You will build the diagnostics, monitoring and dashboards that make fleet health visible, from MCU and SoC firmware to fleet dashboards, and you will bring our hardware to life in vehicles and support the path to production-line testing.
What the Candidate Will Do
• Diagnostics and fault management: Design and implement fault detection, event and DTC handling, UDS and DoIP services, and watchdog supervision across the MCU and the Linux compute platform.
• Monitoring framework: Design and own the vehicle health monitoring pipeline: collect metrics, logs and fault events from the MCU (CAN, serial or UDP) and the Linux SoC, and send them to a central time-series and log store using standard tooling (for example Prometheus exporters or OpenTelemetry collectors).
• Deep system observability: Use eBPF and kernel tracing to observe Linux SoC behavior with low overhead: scheduling, network, driver and camera pipeline latency, and frame or packet drops.
• Dashboards and alerting: Build dashboards (for example in Grafana) for fleet, vehicle and board health: temperatures, voltages, power state, bus load, time sync offset, camera link status and update status. Define alert rules and thresholds, and route alerts to on-call.
• Vehicle bring-up and integration: Commission the compute, MCU and sensor stack in vehicles (CAN-FD, Ethernet, power sequencing, GNSS, IMU, cameras) and validate gPTP and PPS timing.
• Field triage and test automation: Use monitoring data to reproduce and root-cause vehicle issues with the reliability and software teams, and turn each one into an automated regression test and, where useful, a new dashboard panel or alert.
• Safety and security of diagnostics: Contribute to safety analysis (FMEA, FMEDA) and implement safety mechanisms. Protect diagnostic access with authenticated sessions and security access.
• Production line support: Define the diagnostic interfaces and readouts (unit identity, firmware versions, pass or fail data) that end-of-line test stations use, with the manufacturing test team.
Basic Qualifications
• Embedded experience: Bachelor's degree in Electrical Engineering, Computer Engineering, Computer Science or a related field, with 7+ years developing embedded software in C/C++ on an RTOS, embedded Linux or bare metal.
• Automotive diagnostics: Hands-on experience with UDS (ISO 14229), DTC handling and ISO-TP over CAN; DoIP (ISO 13400) is a plus.
• Vehicle networks and integration: Strong experience with CAN/CAN-FD and Automotive Ethernet, tools such as Vector CANoe or similar, oscilloscopes and logic analyzers, and integrating compute platforms and sensors in vehicles.
• Linux observability: Hands-on experience debugging system-level issues on embedded Linux with eBPF-based tools and profilers (for example bpftrace, bcc or perf).
• Monitoring and visualization: Experience building monitoring for devices or fleets (for example with Prometheus and Grafana): instrumentation or exporters, dashboards and alert rules, in Python, Go or C/C++.
• Production lifecycle: Experience taking embedded systems from prototype and third-party integration to production, ideally in autonomous vehicle, robotics or mission-critical automotive domains.
Preferred Qualifications
• Observability stack: Experience with log and time-series stacks (for example Loki, ELK, OpenTelemetry, InfluxDB), eBPF-based observability tools, and fleet-scale telemetry pipelines (retention, cardinality, cost).
• Security: Hardware security module (HSM) and secure boot experience, plus secure diagnostics (UDS security access, authenticated sessions, key management).
• AUTOSAR diagnostics: Configuring Dcm, Dem, FiM, WdgM and E2E modules in an AUTOSAR Classic stack.
• Functional safety: ISO 26262 (ASIL B through D), ASPICE and MISRA C/C++.
• Production and service tooling: End-of-line test, factory programming or remote service diagnostics tooling.
• Linux and AV compute: Linux kernel, device tree and Yocto board support, and experience with platforms such as NVIDIA DRIVE or Jetson.
For Sunnyvale, CA-based roles: The base salary range for this role is USD $180,000 per year - USD $200,000 per year.
You will be eligible to participate in Uber's bonus program, and may be offered an equity award & other types of comp. All full-time employees are eligible to participate in a 401(k) plan. You will also be eligible for various benefits.