About the RoleGreat models are only as good as the system they run on. This role sits at the boundary between the ML autonomy team's models and the on-board system, taking what the research and ML teams build and turning it into something that runs reliably, in real time, on real vehicle hardware. 42dot is looking for builders who want to own the last and hardest mile between a model that works in an offline eval and a model that works in a vehicle.
Founding Team OpportunityThe on-board integration architecture is being defined right now, with no legacy deployment stack to retrofit. An early hire here will help shape:
- The model deployment pipeline, from training-time artifact to vehicle-ready binary
- Inference optimization strategy and tooling for the whole autonomy stack
- The interface contract standard between ML models and on-board software
Responsibilities- Integrate ML models (end-to-end driving, VLA, world models) onto the on-board compute and runtime stack
- Optimize model inference for latency, throughput, memory, and power on embedded automotive hardware
- Build and maintain the model deployment pipeline, including conversion, validation, and rollout
- Stabilize on-vehicle ML execution by diagnosing runtime failures, timing violations, and resource contention
- Define interface contracts between ML models and the rest of the on-board software stack
- Work closely with the ML autonomy team to close the loop between deployment constraints and model design
Qualifications- A track record of grit and ownership, with success moving difficult problems from ambiguity to successful deployment
- Bachelor's, Master's, or PhD in CS, EE, Robotics, or related field
- 5+ years of experience in systems or software engineering, with meaningful time in ML inference, embedded systems, or real-time software
- Experience deploying and optimizing ML models in production, including on resource-constrained hardware
- Strong C++ and Python skills, with experience working across the ML and systems boundary
Preferred Qualifications- Background in autonomous driving or robotics on-board software
- Experience with inference optimization frameworks such as TensorRT, ONNX Runtime, and automotive SoCs
- Experience with real-time operating systems and safety-critical software practices
- Experience building CI/CD and automated validation pipelines
Interview Process- Application Review - Coding Test - 1st interview - 2nd interview - Offer Negotiation - Hiring
- The screening procedures may vary depending on the position, schedule, or other circumstances.
You will be individually notified of the screening schedule and results via the email address provided in your application.
CompensationAdditional Information- In accordance with fair hiring practices, do not include any personal information unrelated to your job qualifications (e.g., Social Security Number, family relations, marital status, age, photo, physical condition, place of birth, etc.) in your resume.
- All documents must be submitted in PDF format and under 30MB in size.
- If you experience issues uploading your resume, please send it along with the job posting URL to [redacted].
- We strongly encourage applications from U.S. veterans and candidates eligible for employment preference under applicable laws.
