Machine Learning Engineer (3D Vision)

FitMatch•New York, NY

115d

About The Position

We are seeking a visionary, highly skilled, and impact-driven Computer Vision Engineer (3D-focused) to join our Data Science team. This is a role designed for a creative problem solver and technical thought leader—someone who looks at complex, high-dimensional data and translates it into real-world human outcomes. In this role, you will own the mission of transforming raw 3D body scan data, measurement features, and body composition metrics into actionable insights. Your work will directly impact cutting-edge healthcare applications and optimize performance tracking for both adolescent and professional athletes in the sports/fitness space. If you want to move beyond passive model training and actively pioneer solutions to real-world physical challenges, this is your sandbox. Note: This position is a contract-to-hire opportunity

Requirements

Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, Applied Mathematics, or a closely related quantitative field.
Minimum of 3+ years of professional experience as a Data Scientist or Machine Learning/Computer Vision Engineer, with a focus on high-dimensional or spatial data domains.
A proven track record of autonomous ownership—taking a model all the way from initial research/prototype to live production deployment.
Expert-level Python and full command of its scientific computing stack.
Deep proficiency with data manipulation libraries like Pandas and NumPy, alongside scientific computing tools like SciPy and Scikit-learn.
High proficiency in Linux/Unix-based terminal interfaces and cloud shell environments (AWS, GCP, or Azure CLI). You should be fully capable of managing compute resources, automating workflows, and troubleshooting cloud infrastructure directly from the command line.
Production experience utilizing PyTorch and/or TensorFlow/Keras.
A strong background in statistical modeling, predictive modeling, and experimental design.
Direct experience handling computer vision tasks relevant to 3D geometry, such as registration, segmentation, and shape analysis.
Strong familiarity with spatial statistics and specialized techniques for analyzing geometric features.

Nice To Haves

Knowledge of cutting-edge geometric deep learning techniques (e.g., PointNet, CNN, DGCNN, GCNs/Graph Neural Networks) for processing irregular 3D structures.
Hands-on experience working with 3D point clouds and/or mesh data structures (e.g., PLY, OBJ, USDZ, STL formats).
Familiarity with modern libraries for geometric processing and visualization, such as Open3D, PCL (Point Cloud Library), or Trimesh.
Scaled deployment experience using Docker and Kubernetes.
Experience with 3D modeling and asset creation in Blender for synthetic data generation or visualization.

Responsibilities

Develop, train, and deploy advanced machine learning and deep learning models for complex spatial analysis of 3D human body scans.
Work directly with ground truth 3D data to creatively build high-performance, robust predictive models that solve tangible physiological pain points.
Synthesize 3D spatial features with multi-modal metadata, including biometrics, demographic information, and self-reported health outcomes to reveal comprehensive health insights.
Design and implement novel algorithms for feature extraction and dimensionality reduction from irregular mesh or point cloud data.
Conduct comprehensive statistical validation and A/B testing of models and deployed features to ensure real-world efficacy and safety.
Act as a technical partner alongside software engineers and domain experts (such as clinicians and biomechanical engineers) to scale and integrate solutions directly into production.
Generate clear, compelling visualizations and reports that effectively communicate nuanced, complex analytical results to both technical teams and non-technical stakeholders.