AI ML Engineer

What we're looking for

We are building a medical imaging device that reconstructs images from noisy, physics-constrained sensor data. This role focuses on
applied machine learning embedded in real systems
, not standalone research models.

You will design, implement, and maintain ML pipelines that integrate with signal processing, hardware, and clinical constraints.

• Develop and train multi-task ML models for stroke detection using RF-derived features, including binary and multi-class classification as well as auxiliary prediction tasks.
• Build and own production-quality ML code for signal → image reconstruction
• Design and evaluate ML approaches for inverse problems under noise and data scarcity
• Integrate ML models with existing signal processing and hardware pipelines
• Debug training failures, data issues, and edge cases
• Make tradeoffs between model complexity, robustness, and interpretability
• Collaborate closely with hardware, DSP, and software engineers

Required

• M.S. in Computer Science, Electrical Engineering, or related field, or 5+ years of industry experience in machine learning
• Experience with generative models (diffusion models, VAEs, or encoder-decoder architectures) applied to 2D/3D data
• Experience building and maintaining ML pipelines beyond notebooks
• Comfort working with imperfect, noisy, real-world data
• Solid foundations in linear algebra, probability, and optimization
• Ability to reason about system-level constraints, not just model performance

• Signal or image processing background, particularly with radar, microwave, or compressed sensing techniques
• Experience with multimodal models or cross-modal generation
• Prior work optimizing models for GPU or edge deployment
• Familiarity with medical device regulatory standards and a clear understanding of the application of AI technologies within FDA-regulated environments.
• Experience in medical device or healthcare technology companies

Relevant ML Experience

Experience with one or more of the following
in applied contexts
:

• Inverse problems / reconstruction
• Generative or probabilistic models (e.g., VAEs, diffusion, GANs)
• CNNs or learned image reconstruction
• Physics-informed or hybrid signal-processing + ML approaches
• Uncertainty estimation and robustness

We care about how you choose models, not whether you’ve memorized architectures.