AI Imaging Engineer for Vascular Diagnostics

Philips; Bedford (MA); On-site; USD 129,000 - 205,000; Full time

Job summary

Philips in Bedford, MA is seeking an AI / Algorithm Engineer to advance vascular imaging algorithms from prototyping to medical product readiness, including X-ray angiography and OCT/IVUS. This office-based, full-time role requires 2+ years in ML or software within regulated devices and strong Python/TensorFlow skills for training, inference, and GPU optimization.

The role involves porting and optimizing pipelines, cross-framework model conversions, verification, and regulatory submissions, with

Qualifications

• 2+ years of post-graduation experience in algorithms, ML, or software development.

Responsibilities

• Port and optimize deep learning pipelines across hardware, improving memory, throughput, and latency.
• Convert and integrate models across frameworks and runtimes (e.g., ONNX) with robust product code.
• Build automated tests and benchmarks to verify ported algorithms against references.
• Develop new algorithms for X-ray angiography and intravascular imaging through prototyping to production.
• Drive data sourcing, curation, and annotation, and evaluate model performance with statistics.
• Contribute to verification, validation, risk management, and regulatory submissions for AI-enabled medical devices.

Skills

Python

Deep Learning

Image Processing

GPU Optimization

Software Testing

Education

Bachelor's degree in Computer Science / Electrical Engineering / Biomedical Engineering

Tools

TensorFlow

PyTorch

ONNX

CUDA

CI/CD

Git

Job description

Philips in Bedford, MA is seeking an AI / Algorithm Engineer to advance vascular imaging algorithms from prototyping to medical product readiness, including X-ray angiography and OCT/IVUS. This office-based, full-time role requires 2+ years in ML or software within regulated devices and strong Python/TensorFlow skills for training, inference, and GPU optimization.

The role involves porting and optimizing pipelines, cross-framework model conversions, verification, and regulatory submissions, with