Adas Feature Engineer

Wayve Wayve · Robotics · Leonberg, Germany · Product & Delivery

This role focuses on building the application-layer software that integrates Wayve's AI capabilities into real vehicle behavior for ADAS features. It involves designing and implementing C++ feature logic, validation tools, and system behaviors, working closely with ML, product, and vehicle integration teams to translate model outputs into reliable and customer-relevant ADAS features. The role emphasizes real-world testing, simulation, and debugging to ensure feature performance and robustness.

What you'd actually do

  1. Design, implement, and maintain C++ application software for ADAS and active-safety-related vehicle features.
  2. Build feature-level logic on top of AI / ML outputs, including validation, feasibility checks, state machines, fallback behaviours, and safety-aware decision logic.
  3. Work with ML engineers to understand model outputs, limitations, failure modes, and how these translate into vehicle behaviour.
  4. Use logs, simulation, replay, and vehicle testing to debug, tune, and validate feature behaviour.
  5. Define and improve metrics, test cases, and validation strategies for ADAS feature performance, robustness, and quality.

Skills

Required

  • Strong C++ software engineering experience
  • Hands-on experience in ADAS, autonomous driving, robotics, vehicle software, active safety, or closely related domains
  • Practical understanding of vehicle feature development
  • Ability to reason about vehicle behaviour, sensor/model inputs, timing, failure modes, and feature-level decision logic
  • Experience working cross-functionally with teams such as ML, perception, planning, controls, vehicle integration, product, or systems engineering
  • Strong problem-solving skills
  • Pragmatic engineering trade-offs under ambiguity
  • Quality mindset
  • Experience writing testable, maintainable software
  • Using data to validate behaviour

Nice to have

  • Experience with ADAS features such as AEB, ISA, AES, ACC, lane keeping, collision avoidance, trajectory validation, or active safety systems
  • Experience at an automotive OEM, Tier 1 supplier, autonomous driving company, robotics company, or vehicle technology startup
  • Familiarity with ML or AI-based autonomy systems
  • Experience with ROS, Linux, Bazel, CMake, Docker, QNX, protobuf, MCAP, CAN, calibration, or vehicle logging systems
  • Experience with vehicle test tracks, public-road testing, HIL/SIL, scenario-based testing, or NCAP-style validation
  • Familiarity with tools used in automotive development and testing, such as CANoe, Vector tools, MicroAutoBox, or similar

What the JD emphasized

  • production or safety-relevant systems
  • ADAS, autonomous driving, robotics, vehicle software, active safety, or closely related domains
  • real-world testing, simulation, replay, logs, or prototype vehicle debugging
  • reason about vehicle behaviour, sensor/model inputs, timing, failure modes, and feature-level decision logic
  • cross-functionally with teams such as ML, perception, planning, controls, vehicle integration, product, or systems engineering
  • pragmatic engineering trade-offs under ambiguity
  • quality mindset
  • testable, maintainable software
  • using data to validate behaviour

Other signals

  • AI software for vehicles
  • feature logic on top of AI/ML outputs
  • customer-relevant ADAS features