Audio ML Engineer
Description
Senior AI/ML Engineer – Audio | Munich
Hear what the camera can’t.
When an airbag clip clicks into place on an assembly line, the camera can’t see it because the airbag is in the way. You’ll build the system that hears it.
Our client is a Munich-based Physical AI startup. Workers at major OEMs wear its smart glasses on the assembly line, and its AI assistant runs on them. Your job is to make audio a live production check: pick out the one click that matters and block out everything else, in real time, at the edge.
What you’ll do
- Build models that detect short, critical sounds (clicks, latches, seating confirmations) on loud factory floors
- Filter and trim audio on the device before it reaches the model, so responses stay real-time and compute isn’t wasted on noise
- Hit hard recall and false-positive targets in client acceptance tests
- Combine audio with vision, IMU and other sensor data
- Optimise models for constrained edge hardware
What you bring
- Strong PyTorch and production ML experience
- Hands-on work in sound event detection, acoustic anomaly detection, noise suppression or real-time audio DSP
- Models deployed on edge or embedded hardware (ONNX, TensorRT, quantisation)
- Bonus: other sensor data (IMU, sensor fusion), manufacturing or robotics, synthetic data or audio augmentation
Why join
Your models go onto real assembly lines within weeks. You’ll work as a peer to the Lead ML Engineer, with a direct line to the CTO. Strong equity opportunity.
Skills
Sign in to read and join the discussion.