Top Tech Jobs & Startup Jobs in Austin, TX

Architect GPU boot, PCIe/CXL enumeration, and capability discovery infrastructure. Coordinate feature planning with hardware, software, and verification teams; develop C++ functional and performance models, create validation plans, and support post-silicon activities. The role requires deep computer and system architecture expertise, production C/C++ experience, and knowledge of advanced interconnects, security constructs, RISC-V, and microcode.

Owns end-to-end architecture and implementation of high-performance communication and memory-management libraries for distributed AI. The role drives GPU, DPU, NIC, and switch hardware-software co-optimization; profiles data movement across memory, storage, and networks; integrates networking into AI serving frameworks; contributes to open source; mentors engineers; leads design reviews; and prototypes emerging technologies.

The role involves developing test plans, implementing verification infrastructures, and collaborating with design teams to ensure the correctness of global IP across multiple products.

Develop AI-driven power analysis and optimization methods for NVIDIA GPUs and SoCs. Analyze RTL and gate-level power, build machine learning, reinforcement learning, and LLM-based models, automate power-analysis workflows, identify anomalies and root causes, and recommend architectural and microarchitectural improvements. Collaborate with architecture, ASIC design, physical design, performance, and software teams while prototyping power-conscious features in Verilog.

Design, build, and maintain infrastructure software for large-scale AI training and inference. Improve efficiency, resiliency, availability, observability, and reliability across distributed AI systems. Develop APIs and tools, analyze failures from application through hardware levels, define reliability metrics, and support NVIDIA AI platforms. The role requires expertise in distributed systems, monitoring, debugging, software engineering practices, and AI infrastructure, with preferred experience in large clusters, RDMA stacks, telemetry, and deep learning frameworks.

Conduct research on efficient, scalable, resilient, and trustworthy machine learning systems. Develop hardware, software, and infrastructure solutions; co-design AI algorithms and systems; analyze performance, scaling, and resilience; collaborate across hardware, software, AI, and networking teams; and publish original research.

Responsible for physical design and implementation of GPUs and ASICs, including methodology establishment, power/clock distribution, and timing closure.

Verify GPU micro-architecture using formal verification tools: define scope, write and implement formal testplans (assumptions, assertions, covers), develop abstractions to obtain proofs, drive tool performance, debug RTL, create scripts/flows, and communicate coverage to partners.

Verify NVIDIA system-on-chip designs by developing unit and subsystem verification environments, test plans, UVM components, sequences, tests, scoreboards, and functional coverage. Run regressions, debug issues, automate build and triage workflows, and improve verification efficiency. Collaborate with architects, designers, and software engineers to meet design verification milestones and ensure high-quality RTL and system-level IP.

Own and advance RTL-to-GDS physical design methodologies for advanced-node GPU, CPU, and SoC designs. Collaborate with P&R engine developers, define scalable flows, run large-scale experiments, analyze QoR data using Python and ML/GenAI, resolve timing and performance issues, and drive PPA, schedule, tool adoption, and methodology improvements across design, CAD, and R&D teams.