Audio ML Engineer

• Develop, optimize, and deploy audio ML models across CPUs, GPUs, and NPUs. 
• Perform model evaluation, quantization, compression, and optimization for embedded deployment. 
• Integrate ML models into real-time audio pipelines such as ECNS and speech enhancement. 
• Analyze and benchmark model architectures including CNNs, RNNs, Transformers, U-Nets, diffusion and sequence models. 
• Optimize inference for low latency, power efficiency, and memory footprint. 
• Profile and debug model performance on embedded hardware platforms. 
• Collaborate with ML, DSP, and systems teams to balance accuracy vs efficiency trade-offs. 
• Support end-to-end deployment, validation, and productization of models. 

Minimum Qualifications 

• Strong programming skills in C/C++ and Python. 
• 3+ years of experience in ML model development, deployment, or embedded systems. 
• Hands-on experience with model quantization and optimization. 
• Strong understanding of real-time system constraints (latency, memory, power). 
• Experience with embedded systems, DSP or AI accelerators. 
• Familiarity with ML frameworks such as PyTorch, TensorFlow, ONNX. 

Preferred Qualifications 

• Experience with audio/speech ML models (speech enhancement, denoising, ECNS). 
• Experience with DSP + ML hybrid systems. 
• Knowledge of Qualcomm AI stack including QNN and Snapdragon platforms. 
• Experience with model optimization techniques such as pruning, compression, and quantization-aware training. 
• Strong understanding of audio processing pipelines and system-level integration. 
• Experience with multi-threading, embedded software design, and real-time debugging on any embedded platform