Edge ML Ops Engineer: Model Conversion & Deployment

torcrobotics; United States; On-site; USD 139,000 - 167,000; Full time

Benefits offered by this job

Bonus

Stock options

Medical premiums

401K with match

Flexible schedule

Paid vacation

Job summary

Torc Robotics is seeking a Software Engineer for the ML Ops Framework & Conversion team. You will implement model conversion and deployment pipelines for edge devices such as NVIDIA Orin, and manage the model release registry with strong focus on latency and accuracy.

You will work with perception and safety teams to ensure models meet strict performance targets, validating across platforms and collaborating with stakeholders to deliver reliable results.

Qualifications

• Experience with model conversion and deployment pipelines.
• Ability to benchmark latency and ensure parity across platforms.
• Experience managing a production model release registry.

Responsibilities

• Architect and implement model conversion and compilation pipelines.
• Maintain and evolve the model release registry across platforms.
• Conduct latency benchmarking and quality parity validation.
• Compare metrics across hardware to ensure accuracy and latency targets.
• Collaborate with model developers and stakeholders to drive reliable outcomes.

Skills

Model conversion

Latency benchmarking

Edge deployment

Model registry

Cross-platform metrics

Education

Bachelor's degree in CS/EE/Robotics or related

Tools

ONNX

TensorRT

torch.compile

NVIDIA Orin

Job description

Torc Robotics is seeking a Software Engineer for the ML Ops Framework & Conversion team. You will implement model conversion and deployment pipelines for edge devices such as NVIDIA Orin, and manage the model release registry with strong focus on latency and accuracy.

You will work with perception and safety teams to ensure models meet strict performance targets, validating across platforms and collaborating with stakeholders to deliver reliable results.