Real-Time Computational Imaging Engineer

Merge; San Francisco, Northern (CA, KY); Hybrid; USD 180,000 - 240,000; Full time

Job summary

Merge Labs in San Francisco is seeking an experienced engineer to build and maintain a shared imaging core that turns raw ultrasound data into image series across our platforms. You will define performance objectives and develop, optimize, and validate reconstruction methods for real-time and offline use.

You will work with scientists and engineers to push imaging throughput within compute and memory budgets, explore learned methods, and ensure rigorous validation and clear reporting of

Qualifications

• Strong signal-processing and computational-imaging fundamentals.
• Experience developing or materially improving reconstruction methods.
• Strong scientific Python skills and GPU profiling for large data.
• Experience designing and maintaining shared scientific software with tests and docs.
• Rigorous validation with uncertainty reporting.

Responsibilities

• Develop reusable methods for beamforming/backprojection, clutter filtering, and motion estimation within a shared imaging core.
• Optimize imaging pipelines through controlled experiments balancing sensitivity and cost.
• Rigorously validate imaging pipelines with benchmarks, measurements and regression tests.
• Produce reliable imaging results and present findings with explicit quality and limitations.
• Meet throughput and latency targets for real-time and large-scale offline processing.
• Use measured performance to shape device and acquisition requirements.
• Explore learned methods and integrate improvements when proven beneficial.

Skills

Signal processing

Computational imaging

Python

GPU performance

Software design

Tools

Bazel

Job description

Merge Labs in San Francisco is seeking an experienced engineer to build and maintain a shared imaging core that turns raw ultrasound data into image series across our platforms. You will define performance objectives and develop, optimize, and validate reconstruction methods for real-time and offline use.

You will work with scientists and engineers to push imaging throughput within compute and memory budgets, explore learned methods, and ensure rigorous validation and clear reporting of