Radar ML Engineer (Edge Deployment) – TS/SCI Eligible

Blue Origin; Denver (CO); On-site; USD 120,000 - 168,000; Full time

Benefits offered by this job

Relocation assistance

Travel up to 10%

TS/SCI clearance eligibility

Job summary

Blue Origin in Denver is seeking a radar AI/ML engineer to own the AI layer for object detection and characterization from raw radar data. You’ll shape model architecture under constraints of limited labeled data and edge hardware, influencing the autonomy stack and mission outcomes.

You’ll work with RF, sensing, GNC and flight software teams to deploy DL models on embedded systems, with a focus on performance, uncertainty, and robust evaluation for space missions.

Qualifications

• BS in CS/EE/CE/Physics with ~3 years ML engineering experience.
• Experience building/shipping DL models in PyTorch.
• Edge deployment experience (TensorRT, quantization, FPGA/SoC).
• Strong Python skills.
• Radar fundamentals: range, Doppler, angle.

Responsibilities

• Design, train and evaluate DL models (GNNs, transformers, DRL) for radar data.
• Build characterization models for object attributes with calibrated uncertainty.
• Represent radar data for models (points or graphs).
• Develop data engine: synthetic data, simulation training, labeling strategy.
• Detect inputs outside training data and flag low-confidence results.
• Collaborate with systems teams to deploy models on embedded/edge hardware.
• Define metrics, test sets, and evidence for edge deployment.

Skills

Python

PyTorch

Deep learning

Radar fundamentals

Edge deployment

C++

GMTI/AMTI familiarity

Education

Bachelor's degree in CS/EE/CE/Physics

Tools

TensorRT

FPGA

SoC targets

Job description

Blue Origin in Denver is seeking a radar AI/ML engineer to own the AI layer for object detection and characterization from raw radar data. You’ll shape model architecture under constraints of limited labeled data and edge hardware, influencing the autonomy stack and mission outcomes.

You’ll work with RF, sensing, GNC and flight software teams to deploy DL models on embedded systems, with a focus on performance, uncertainty, and robust evaluation for space missions.