Join NVIDIA, a leader in High-Performance Computing and Artificial Intelligence, at the forefront of technological advancement. We are expanding our teams with exceptional talent to drive innovation in AI computing.

We are seeking a Senior System Software Engineer with extensive expertise in speech technologies to support our enterprise and developer customers. This role involves deep technical engagement, focusing on implementing, troubleshooting, and optimizing Automatic Speech Recognition (ASR), Text-to-Speech (TTS), Audio Language Models (ALM), and Speech-to-Speech (S2S) systems in production environments. If you are driven by solving complex conversational AI challenges, we invite you to join our Speech AI Engineering team.

Engage with cutting-edge GPU-accelerated AI systems deployed at scale and tackle complex challenges in real-time streaming audio processing and low-latency inference.

Troubleshoot and resolve intricate issues across ASR, TTS, ALM, and S2S pipelines. Collaborate with Model researchers to transition ASR, TTS, and S2S models from research to production readiness.

Develop and enhance core speech services using C++ and Python backend implementations, leveraging CUDA for GPU acceleration. Optimize inference performance through advanced batching, caching, and multi-threaded pipeline optimizations.

Drive feature development, including advanced voice activity detection, speaker diarization, and decoder implementations. Contribute to Python and C++ client SDKs and CLI tools for seamless service integration.

Assist customers with API integration, SDK usage, model deployment, and performance optimization, providing advanced technical guidance for their speech technology solutions.

A Masters or BE/B.Tech degree in Computer Science, computer architecture, or a related field is required, along with 6+ years of professional experience.

Possess excellent C++ and Python programming and software design skills, including proficiency in debugging, performance analysis, and test design. Experience with inference pipelines for LLMs, Speech Recognition, and Speech Synthesis is essential.

Demonstrate a solid understanding of modern model architectures such as Transformers, CNNs, and RNNs. Exhibit strong debugging abilities across various software layers, including storage systems, kernels, and containers.

Experience building and deploying cloud services using HTTP REST, gRPC, and Websockets is necessary. Strong collaborative and interpersonal skills, with a proven ability to influence within a dynamic matrix environment, are crucial.

Ability to work independently, define project goals, and manage your own development efforts is expected. Knowledge of real-time streaming audio systems and low-latency architectures, along with experience in speech model fine-tuning or customization, will set candidates apart. Publications or contributions to ML optimization open-source projects, or experience with embedded systems or edge deployment, are considered advantageous.