Signal Processing and ML Engineer

Are you passionate about the intersection of classical signal processing and deep learning? Do you want to use raw radar data to define the future of sports metrics?

Trackman is looking for a Signal Processing and ML Engineer to become one of our key engineers in evolving our radar tracking system in everything from traditional real time signal processing to advanced neural network-based architectures. If you are an experienced Signal Processing engineer who loves building production-grade solutions, we want to hear from you.

Trackman has led the world in sports tracking technology since 2003. We develop innovative solutions that integrate computer vision, radar technologies, and advanced mathematical models. Our products are used by the world's top athletes, and the solution must meet the highest standards of accuracy, performance, and reliability.

The Role

As a Signal Processing and ML Engineer, you will work across the entire radar tracking stack from signal acquisition and real-time tracking algorithms to Python analysis tooling and ML model deployment. You will have full ownership of the solution from design to final release. Additionally, you will conduct statistical analysis to verify performance, identify subtle biases, and ensure our solutions deliver industry-leading accuracy across all conditions.

The Team

You will be a part of a R&D team called Golf Launch Monitors. The team is responsible for designing, building, and maintaining high-quality golf launch monitor solutions. The team consists of specialists in radar signal processing, computer vision, mathematical modeling, and software engineering. The team is committed to excellence, where we constantly push boundaries to ensure our performance stays ahead of the curve.

Key Responsibilities

• Real-time signal processing: FFT pipelines, noise floor estimation, peak detection, tracking filters, beamforming, ambiguity resolution among other things.

• Algorithm Design: Develop and optimize signal processing chains that combine classical techniques with neural network architectures for feature detection.

• Data-Driven Discovery: Use statistical methods to identify edge cases in real-world sports data and refine our models to handle complex environmental scenarios.

• Statistical Validation: Perform deep-dive statistical analysis on massive datasets to validate accuracy and ensure our products meet strict performance benchmarks post-release.

• Collaborative Engineering: Partner with other teams on everything related to hardware and sensor performance, software structure and deployment and proper QA testing.