Architect - GPU Performance

Overview

Architect role focused on GPU and system-on-chip performance analysis and modeling for next-generation visual, automotive, and HPC systems.

What you'll do

• Perform system-level and bottleneck analysis for complex, high-performance GPUs and SoCs.
• Build hardware models across abstraction levels, including performance models, RTL test benches, emulators, and silicon-based analysis.
• Develop workloads and test suites for graphics, machine learning, automotive, video, and compute vision applications.
• Collaborate on architecture trade-offs affecting system performance, area, and power consumption.
• Develop infrastructure such as performance models, testbench components, and performance analysis/visualization tools.
• Drive methodologies to improve turnaround time and enable early performance analysis with representative data sets.

What you'll need

• 3+ years of experience with exposure to performance analysis and complex SoC/GPU architectures.
• BE/BTech or MS/MTech in a relevant area (PhD is a plus) or equivalent experience.
• Strong understanding of SoC architecture, graphics pipeline, memory subsystem architecture, and Network-on-Chip (NoC)/interconnect.
• Expert hands-on competence in C/C++ programming and Perl/Python scripting.
• Exposure to Verilog/SystemVerilog and SystemC/TLM is a strong plus.
• Strong debugging and analysis skills, including data/statistical analysis using RTL dumps.
• Hands-on experience developing performance simulators and cycle-accurate/approximate models (strong plus).

Details

• Location: Bengaluru, India.
• Work arrangement indicated: #LI-Hybrid.