What you ll be doing:

• Architect and lead development of next-generation Deep Learning and multimedia algorithms for processing of speech and audio applications.

• Train Speech Enhancement models e.g. Background Noise Reduction, Room Echo Cancellation, Audio Super Resolution etc.

• Analyze model accuracy and bias and recommend the next course of action & Improvements.

• Improve processes for speech data processing, augmentation, filtering & training sets preparation.

• Optimize algorithms for optimal performance on the GPU tensor cores

• Collaborate with various teams to drive an end to end workflow from data curation and training to performance optimization and deployment

• Influence strategic decisions in the team and product roadmap

• Partner with system software engineers and validation teams to build and ship production-quality code.

What we need to see:

• PH.D./MS in Computer Science or a closely related engineering field with 8+ years of relevant experience

• Strong background in Deep Learning including model design, pruning & performance optimization, transfer learning etc

• 4+ years of experience of leading cross-module projects and taking them to productization

• Strong software engineering background with proficiency in C or C++

• Hands-on expertise with PyTorch, TensorRT, CuDNN and one or more Deep Learning frameworks (Tensorflow, Keras etc)

• Familiarity with various DSP algorithms e.g STFT, ISTFT, FFT, Mel and Chroma Spectrograms and audio encoders-decoders e.g. AAC, OPUS is a plus

• CUDA programming experience is a plus

• Excellent communication and collaboration skills

• Self-motivated and able to find creative practical solutions to problems