AIML & Data Engineer (NASA contract) ($120K-$170K)

PHASE3D

AI/ML & Data Engineer (NASA Project)

Full-Time | Individual Contributor | Data Science & AI | US Citizenship Required

About Phase3D

Phase3D is a deep-tech manufacturing intelligence company building Fringe Inspection™ a real-time, structured light-based in-situ inspection system for metal additive manufacturing. Our technology is deployed at customers including multiple aerospace primes, NASA, Air Force, NAVY and other tier 1 customers, and we are growing our engineering team to support rapid product development and commercialization. We hold a NASA SBIR award to advance Born Qualified Additive Manufacturing for space applications. Our technology generates rich, calibrated 3D heightmap datasets from every layer of the metal AM printing process, and we are now building the AI/ML intelligence layer on top of that data.

Role Overview

We are hiring an AI/ML & Data Engineer to lead the development of machine learning models, data pipelines, and scientific feature extraction capabilities for Phase3D's Fringe Inspection™ and Fringe Qualification™ platforms. This is a high-impact individual contributor role at the intersection of physics, data science, and advanced manufacturing.

You will work directly with in-situ inspection data from metal AM machines, structured light heightmaps collected layer-by-layer during live builds, and develop AI/ML models to correlate these signals with post-build defect data from CT scans and other final part quality data. Your work will directly advance Phase3D's NASA SBIR Phase I deliverables and lay the foundation for the Fringe Qualification™ commercial product. This full-time role will be focused on the 2026 SBIR and then transition to other data projects, and is not a temporary contract.

Key Responsibilities

AI/ML Model Development

•      Design, train, and validate machine learning models to correlate in-situ Fringe Inspection™ anomaly data with post-build CT-detected defects (porosity, delamination, surface irregularities)

•      Develop classification and regression models to support go/no-go quality thresholds for metal AM qualification workflows

•      Build and iterate on deep learning and computer vision approaches for anomaly detection in 3D heightmap data

•      Implement uncertainty quantification and confidence interval frameworks to characterize model reliability for sensor-based predictions

Scientific Feature Extraction

•      Extract physically meaningful features from structured light heightmap data, including surface roughness metrics, volumetric anomaly measurements, layer-to-layer deviation analysis, and defect geometry characterization

•      Develop signal processing pipelines to denoise, normalize, and condition raw sensor data for model input

•      Collaborate with the engineering team to encode domain knowledge (AM physics, material behavior, process parameters) into feature engineering and model architecture decisions

•      Translate scientific outputs into actionable thresholds and indicators for use in Fringe Qualification™ enterprise dashboards

Large Dataset Management & Infrastructure

•      Architect and maintain scalable data pipelines to ingest, store, and process large volumes of layerwise inspection data (targeting 500,000+ labeled layers across multiple machines, materials, and build conditions)

•      Build and manage coupled datasets pairing in-situ Fringe Inspection™ data with corresponding CT scan ground truth for supervised learning

•      Establish data versioning, labeling standards, and traceability protocols to support NASA qualification requirements (NASA-STD-6030, NASA-STD-6033)

•      Design and maintain cloud-connected, CCMC-compliant storage and processing infrastructure consistent with Phase3D's existing data systems

Commercial Feature Development

•      Work with the product and commercial teams to translate ML insights into customer-facing features within Fringe Qualification™, including automated qualification reports, anomaly trend dashboards, and threshold recommendation tools

•      Develop subscription-ready datasets with statistical confidence intervals and material-specific allowable thresholds for use by aerospace and defense customers

•      Support Phase3D's vision of a machine-agnostic qualification platform by building models that generalize across printer types (EOS M290, M460, M300-4), materials (Ti-64, GRCop, Invar 36), and facilities

NASA SBIR Research Execution

•      Lead statistical analysis and correlation studies between in-situ anomaly data and post-build CT defect results, supporting Phase I deliverables for Phase3D's NASA SBIR award (Topic I01.01: Advanced real-time monitoring and control for additive manufacturing)

•      Develop and validate preliminary go/no-go quality thresholds for Invar 36 at our partner facility’s EOS M300-4 platform

•      Contribute to Phase I final report and Phase II roadmap, including documentation of anomaly-defect correlation methodology, model performance metrics, and threshold validation approach

•      Support the scientific case for 'Born Qualified' AM by building traceable, defensible, quantitative evidence of in-situ inspection fidelity

What We're Looking For

Required

•      5–8 years of experience in machine learning, data science, or scientific computing — ideally applied to sensor data, imaging, or physical systems

•      Strong proficiency in Python and the ML stack (PyTorch or TensorFlow, scikit-learn, NumPy, pandas, etc.)

•      Experience with large-scale dataset management, data pipeline engineering, and cloud storage systems

•      Demonstrated ability to develop and validate predictive models on real-world, noisy sensor data

•      Solid understanding of statistical methods including hypothesis testing, confidence intervals, and uncertainty quantification

Nice to Have

•      Experience with 3D point cloud, heightmap, or structured light data processing

•      Background in non-destructive evaluation (NDE), manufacturing quality control, or materials science

•      Familiarity with CT scan data and defect characterization in metals

•      Experience working under government R&D contracts (SBIR, STTR, DOE, AFRL, NASA)

•      Knowledge of additive manufacturing processes (LPBF/SLM preferred)

•      Publications or contributions in applied ML, manufacturing intelligence, or scientific data analysis

What We Offer

•      Direct ownership of Phase3D's AI/ML capabilities, a greenfield opportunity to build from the ground up

•      Access to unique, proprietary datasets from our data collection deployments

•      The opportunity to contribute to NASA-funded research with real-world aerospace qualification impact

•      Competitive compensation including meaningful equity participation

•      A fast-moving, technically deep team that values scientific rigor and ships real products

Phase3D is an equal opportunity employer. We celebrate diversity and are committed to creating an inclusive environment for all employees.