Architect — AI-Powered Performance Verification Automation
NVIDIAJob Title
Architect — AI-Powered Performance Verification Automation
Role Summary
Design, develop, and deploy AI/ML-driven automation to accelerate hardware performance verification for GPU and SoC projects. Work with verification teams to automate test generation, triage, analysis, and integrate intelligent assistants into verification workflows.
Experience Level
Senior; typically requires 3+ years in hardware verification, performance validation, or EDA tool development.
Responsibilities
Primary responsibilities include:
- Build AI/ML tools to generate, triage, and analyze performance test cases and results.
- Develop intelligent agents to detect performance regressions, determine root causes, and recommend corrective actions.
- Integrate LLM-based assistants into verification infrastructure for natural-language queries of results, specifications, and coverage data.
- Design data pipelines to collect, curate, and label verification data for model training and continuous improvement.
- Collaborate with verification engineers to define automation priorities and validate AI-driven solutions in real workflows.
- Establish metrics and dashboards to measure automation impact (cycle time reduction, coverage improvement, engineer productivity).
- Stay current with generative AI, reinforcement learning, and formal methods as they apply to hardware verification.
Requirements
Must-have qualifications followed by useful additions.
- 3+ years experience in hardware verification, performance validation, or EDA tool development.
- Strong programming skills in Python.
- Hands-on experience with ML/AI frameworks (PyTorch, TensorFlow, scikit-learn) or LLM APIs (OpenAI, NVIDIA NIM/NeMo).
- Understanding of performance verification methodologies (benchmarking, profiling, regression analysis).
- Experience with CI/CD pipelines and infrastructure automation.
- Nice-to-have: Familiarity with C/C++ and SystemVerilog/UVM.
- Nice-to-have: Experience applying ML to EDA or verification problems (test generation, bug prediction, coverage closure).
- Nice-to-have: Familiarity with RAG architectures, prompt engineering, and agent frameworks (LangChain, CrewAI).
- Nice-to-have: Knowledge of GPU/SoC architecture or published work/patents in related domains.
Education Requirements
B.Tech, M.Tech, or PhD in Electrical Engineering, Computer Science, or a related field.
About the Company
Company: NVIDIA
Headquarters: Santa Clara, California, USA
NVIDIA is a global leader in accelerated computing, renowned for its innovative solutions in AI and digital twins that transform diverse industries. The company specializes in networking technologies, providing end-to-end InfiniBand and Ethernet solutions for servers and storage that optimize performance and scalability. NVIDIA serves sectors such as high-performance computing, enterprise data centers, and cloud computing, constantly reinventing its products and services to stay ahead in the market.
