Top Tech Jobs & Startup Jobs in Pune

Lead and grow a team building an on-device AI inference platform for RTX/ DGX GPUs. Drive cross-functional alignment, architect inference runtimes, optimize models and pipelines for low-latency, memory-efficient local deployment, and establish processes for system-level debugging, performance/accuracy trade-offs, and production readiness. Mentor engineers and deliver roadmap-aligned, high-quality releases.

Design and optimize on-device AI inference stacks for RTX and DGX systems. Build inference runtimes, optimize models and pipelines (quantization, pruning), perform system-level debugging and performance-accuracy tradeoffs, and collaborate across software, research, architecture, and product teams to ensure production readiness.

Develop and optimize on-device AI inference stacks and runtimes for RTX and DGX systems. Collaborate with cross-functional teams to design high-performance, memory-efficient inference pipelines, apply model optimizations (quantization, pruning, distillation), perform system-level debugging and performance tuning, and build infrastructure for performance/accuracy evaluation and production readiness.

The role involves developing compilers for NVIDIA GPUs, working on optimization problems, collaborating with a distributed team, and influencing new GPU architectures.

Design and develop low-latency, GPU-accelerated video streaming features; optimize encoder and GPU/CPU pipeline performance; build video quality measurement and analysis tools; integrate AI models for real-time video processing and adaptive streaming; and improve reliability, telemetry, and debugging for cloud-based streaming systems.

Lead research and development of state-of-the-art generative audio and multi-modal deep learning models; define technical roadmap; scale and optimize large foundation models on GPU clusters; develop audio-speech-visual fusion algorithms; productize and deploy optimized inference on NVIDIA GPUs/edge; collaborate across research, hardware, and product teams; mentor engineers and uphold code and design quality.

Architect, develop, and debug system software for NVIDIA Halos across QNX, Linux, and hypervisor environments. Build BSP, boot, drivers, device management, I/O virtualization, and diagnostics for NVIDIA SoCs. Ensure safety and security compliance (ISO 26262/21434), apply formal methods, optimize platform performance, use AI-assisted tools, design tracing/logging solutions, and mentor junior engineers.

Operate and improve the reliability, availability, and performance of large-scale GeForce NOW services. Participate in incident triage and on-call rotations, build automation and tooling, enhance observability (metrics/logs/traces), drive SLO/SRI practices, run postmortems, and design/operate Kubernetes-based services across cloud and datacenter environments.

Design, implement, and operate scalable distributed software and network/system automation for NVIDIA's multi-tenant cloud infrastructure. Own code lifecycle, build maintainable APIs, ensure secure communications and data persistence, contribute to open-source, and present roadmaps to stakeholders.