Machine Learning Applied Researcher - Speech, Vision and Audio
Core
Research and develop frontier deep learning models and algorithms across vision, audio, and sensor modalities to solve unique challenges in automatic speech recognition (ASR) and beyond.
Role type
Applied ML researcher (speech, vision, audio)
Builds
State-of-the-art deep learning models and novel neural network architectures
Domain
Artificial Intelligence / Machine Learning / Speech & Audio
Deliverable
production ML models
Required skills
deep learning theory, PyTorch, large-scale model training, self-supervised learning, synthetic data generation, automatic speech recognition (ASR), multimodal data handling, neural network architecture design
Preferred skills
PhD in CS/EE, publication record at top-tier ML venues, multidisciplinary background (neuroscience, physics, signal processing), experience with distributed teams
Technologies
PyTorch
Responsibilities
Research and develop cutting-edge deep learning models leveraging multiple data modalities; Design and implement novel neural network architectures for ASR; Navigate ambiguous problem spaces by formulating hypotheses and iterating on experiments; Improve model performance while adhering to deployment constraints (latency, memory, compute); Synthesize approaches from different fields into organization-wide solutions; Stay current with scientific literature on frontier ML methods