Research and develop AI-powered audio understanding algorithms for real-time device deployment. Responsibilities include training and evaluating sound event detection, localization, acoustic scene, anomaly detection, and multimodal models; building datasets and evaluation pipelines; optimizing accuracy, latency, and compute cost; collaborating with DSP and software engineers; and contributing to product prototypes, patents, and research publications.

Develop agentic workflows that automate audio machine learning research, including data preparation, training, evaluation, hyperparameter tuning, benchmarking, regression testing, experiment tracking, and reporting. Build reliable tools for GPU scheduling, artifact management, reproducible execution, and LLM-agent integration with research resources. Collaborate with audio researchers to create reusable workflows, dashboards, and reports for real-time audio processing algorithms.