NVIDIA SOC Architecture team is looking for a Senior Data Scientist with SW development skills and HW-System architecture experience. You'll be part of the Artificial Intelligence Revolution, working with world-class systems architects and deep learning experts to define the next generation SoC. NVIDIA is developing processor and system architectures at the forefront of accelerating machine learning, automotive and high-performance computing applications, building the most advanced SoCs in the world.
In this role, you will:
- Develop datasets, AI Models, and train AI models for advanced system architecture Power and Performance features
- Explore and define future aspects of architectures that bring together NVIDIA GPUs, custom processors, and accelerators into a single chip
- Build performance models, automate analysis workflows, and generate data for pre and post silicon characterization
- Work with HW & System architects, software, ASIC design, verification, physical design, VLSI, and platform teams
- Perform performance, perf @ watt, and power modeling and analysis to optimize architecture
- Develop and validate System C/C++ models for early SW development and verification
- Drive automation & tools for architectural exploration and analysis
- Provide post-Silicon production support with silicon debug and analysis for performance and power optimization
- Collaborate with SoC Architects to initiate efficiency improvements
- Analyze the interplay of hardware and software architectures on future algorithms and applications
Requirements:
- B.Sc., M.Sc. or Ph.D. in Computer Science, Electrical Engineering, or related field
- 5+ years of relevant experience
- Programming proficiency in Python, System C/C++
- Familiarity with Linux development environment
- Strong interpersonal and organizational skills
- Excellent analytical and verbal communication skills
- Independence and drive to lead initiatives
Preferred qualifications:
- Data scientist experience
- Background in Power and Performance model development
- Experience in architecture workflows automation and tooling
- Knowledge in accelerated computing, machine learning, NLP, AIGC, LLM, AI4S
- Experience with CUDA and deep learning frameworks (TensorFlow or PyTorch)
- Proficiency with Python data analysis packages like Pandas and NumPy
NVIDIA is an equal opportunity employer committed to fostering a diverse work environment.