Data Scientist - Predictive Manufacturing & Geometry ML

Hadrian Automation; Torrance (CA); On-site; USD 170,000 - 300,000; Full time

Benefits offered by this job

Health insurance

Dental insurance

Vision insurance

Life insurance

401(k)

Relocation support

Flexible vacation

Equity

Job summary

Hadrian is building autonomous factories with AI, software, robotics, and full-stack manufacturing to transform aircraft, space, and defense sectors in the U.S. This role focuses on predictive manufacturing data science, turning CAD geometry, material, tolerances, and routing into accurate cycle-time and cost predictions.

You’ll own pipelines—from feature engineering to deployment—enabling better quoting, scheduling, and capacity decisions.

Qualifications

• Experience forecasting on real manufacturing data with honest uncertainty.
• Proficient in Python and ML frameworks used in production.
• Experience with backtesting, leakage control, and calibration.

Responsibilities

• Build and ship production models for cycle time, tool life, quality, and demand.
• Model directly off geometry by engineering features and geometric/graph models.
• Create part and operation embeddings to transfer behavior to cold-start parts.
• Validate with backtesting that respects time ordering and guards against leakage.
• Own models end to end on the platform, including training, serving, monitoring, and retraining.
• Detect drift and quality anomalies to improve predictions as data lands.
• Turn predictions into decisions for quoting, scheduling, capacity, and DFM; design experiments and AB tests.

Skills

Forecasting

Representation learning

Deep learning

Classical ML

Validation

Python

Deployment & monitoring

Tools

PyTorch

PyTorch Geometric

CAD features

Job description

Hadrian is building autonomous factories with AI, software, robotics, and full-stack manufacturing to transform aircraft, space, and defense sectors in the U.S. This role focuses on predictive manufacturing data science, turning CAD geometry, material, tolerances, and routing into accurate cycle-time and cost predictions.

You’ll own pipelines—from feature engineering to deployment—enabling better quoting, scheduling, and capacity decisions.