Top Tech Jobs & Startup Jobs

Verify ASIC/GPU and SoC designs using advanced verification methodologies. Define verification scope, develop test plans and test generators, build verification infrastructure, and collaborate with architects, designers, and pre/post-silicon teams to validate CPU/GPU interactions and memory coherence.

Design and implement production-grade VLSI/EDA tools and methodologies for multi-die package design. Build flows, applications, and automation to construct, analyze, validate, and optimize package designs to accelerate tape-out and scale engineering teams.

Design and implement reliable, scalable full‑stack web tools (backend and frontend) to automate and optimize chip design and verification workflows. Build web applications in Go, TypeScript, and JavaScript, model hardware-level product definitions, research solutions to improve engineering efficiency, and support hundreds of global users.

Design, develop, and verify Tegra core boot firmware for safety- and security-critical platforms. Define and decompose functional/safety requirements, perform FMEA/DFA on bootloader features, validate on pre-silicon simulation platforms, support silicon bring-up and post-silicon issues, and collaborate with cross-functional teams to meet automotive and datacenter quality and verification standards.

Validate clocking and reset logic for Tegra and GPU ASICs end-to-end: develop test plans, automation and validation flows, implement coverage metrics, run tests, identify/fix bugs, partner with architecture/design teams, and productize verification solutions across platforms.

Lead performance characterization and optimization of large-scale AI workloads across GPUs, networking, storage, and software. Design and run benchmarks and profiling, diagnose bottlenecks, define evaluation methodologies and metrics, produce actionable optimization plans, and collaborate with DL engineers, platform teams, and GPU architects to deliver improvements.

Lead development of local AI agent frameworks and runtimes on Windows and GeForce RTX GPUs. Optimize agent inference, sandboxing, and privacy/security policies, collaborate with AI research and driver teams, mentor engineers, and deliver production-ready, local autonomous agent systems.

Develop and maintain open-source GPU-accelerated inference software for LLMs. Enable disaggregated serving across vLLM, SGLang, and TensorRT-LLM for agentic inference (long-horizon reasoning, tool calling, multi-turn state). Optimize inference-state management, GPU memory and cache reuse, and distributed frontend performance, while prioritizing latency, throughput, and integration with upstream APIs and agent harnesses.

Deploy, manage, validate, and optimize large-scale AI compute and HPC infrastructure in Linux environments. Serve as a technical liaison among NVIDIA, partners, and enterprise customers; lead planning through implementation, document handoffs, transfer knowledge, troubleshoot hardware and software, and provide feedback for product and architecture improvements. The role involves networking, system design, automation, benchmarking, cluster provisioning, and customer-site travel up to 30%.

Design and build RL post-training infrastructure for AI, optimizing training-inference-rollout on distributed systems while ensuring fault tolerance and scalability.