Software R&D Engineer, VLSI Physical Design - New College Grad 2026
NVIDIAJob Title
Software R&D Engineer, VLSI Physical Design (New College Grad 2026)
Role Summary
Join NVIDIA's VLSI physical design software team to develop high-performance optimization engines and tools used in advanced chip design. The team combines parallel computing, algorithmic research, and machine learning to improve placement, timing, power, and routing for AI hardware.
This role focuses on implementing and improving algorithms for physical-design flows, working closely with design teams to deploy solutions that raise frequency and reduce power on advanced process nodes.
Experience Level
Entry-level (New college graduate). Suitable for new graduates and candidates with up to ~2 years of relevant experience.
Responsibilities
Primary responsibilities include research, implementation, and deployment of physical-design software and algorithms.
- Invent and implement optimization engines that combine traditionally separate steps (e.g., legalization, sizing, buffering) to improve PPA (performance, power, area).
- Design, optimize, and maintain C++ algorithms for gate-level sizing, buffering, useful clock skew, cell legalization, power minimization, ECO routing, and incremental extraction.
- Profile and improve performance: multithreading, memory and I/O efficiency, and scalable compute usage.
- Work with design teams to validate and integrate tools into production flows and assist with deployment on real designs.
- Investigate and apply machine-learning techniques (e.g., RL, GNNs) where they provide measurable improvement to physical-design problems.
Requirements
Must-have technical skills and practical experience for day-one impact. Nice-to-have items indicate preferred additional strengths.
- Must-have: Strong C++ programming skills and experience developing algorithms or software (production-quality C++ coding).
- Must-have: Understanding of VLSI timing optimization and related concepts: cell libraries, interconnect models, crosstalk, glitches, IR drop, timing constraints, corners, and congestion.
- Must-have: Practical experience with physical-design tools and flows (examples: ICC2, Innovus, PrimeTime, Tempus, StarRC) and scripting (Perl, Tcl, Python).
- Nice-to-have: Familiarity with modern C++ (C++14+ features such as lambdas and concurrency), multithreading and high-performance software design.
- Nice-to-have: Experience applying machine learning (reinforcement learning, GNNs) to physical-design or related optimization problems.
- Nice-to-have: Demonstrated ability to fuse multiple physical-design steps into hybrid optimization engines.
Education Requirements
Master's or PhD in Electrical Engineering or Computer Science preferred; equivalent practical experience accepted.
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.
