Mid-Level ML Engineer, Predictive Quality for Additive Manufacturing

The challenge

Additive manufactured parts have a 12% scrap rate due to porosity defects that are only caught during post-build CT inspection. You will build in-situ monitoring models that detect porosity formation during the build process.

Role overview

In-situ sensor data -- melt pool imagery, pyrometer readings, layer-by-layer scans -- fed into ML models that predict part quality during metal additive manufacturing. The goal is detecting defect formation as it happens, turning post-build CT inspection into real-time process control.

Specializations

Technical requirements

• MS in Materials Science, Mechanical Engineering, or Computer Science with ML focus
• 3-5 years applying ML to manufacturing or materials science problems
• Experience with image-based ML (CNNs for spatial feature extraction)
• Proficiency in Python, PyTorch or TensorFlow, and data processing with pandas/NumPy
• Understanding of metal additive manufacturing processes (L-PBF preferred)
• Experience with multi-modal sensor data fusion

Key responsibilities

• Develop ML models for real-time defect detection during metal powder bed fusion processes
• Process and analyze in-situ monitoring data: melt pool imagery, pyrometer signals, and recoater force measurements
• Build process-structure linkage models that predict porosity, lack of fusion, and keyhole defects from process parameters
• Design data pipelines that ingest multi-modal sensor data from 20+ AM machines in near real-time
• Validate model predictions against post-build CT inspection results
• Collaborate with process engineers to define actionable thresholds for build intervention decisions

Compensation & Benefits

• Base salary: $120,000 - $155,000
• Annual bonus 8%
• Tuition support for PhD candidates
• Access to state-of-the-art AM lab facilities
• 4 weeks PTO