Senior Applied Research Engineer, Accelerator Algorithms
Core
Drive applied research and architecture for algorithms on NVIDIA's Programmable Vision Accelerator (PVA) to optimize real-world autonomous vehicle and physical AI workloads.
Role type
Senior Applied Research Engineer (Accelerator Algorithms)
Builds
Optimized algorithms, prototypes, benchmarks, and performance models for PVA hardware and software stacks.
Domain
Accelerated computing, autonomous vehicles, robotics, physical AI, computer architecture.
Deliverable
production ML models
Required skills
Workload characterization, instruction-level parallelism, memory-aware scheduling, vectorized execution, HW/SW co-design, performance modeling, C++, CUDA, Python, VLIW/SIMD, fixed-point arithmetic, sensor processing, multimodal perception.
Preferred skills
ROS/ROS2, AV/robotics middleware, safety-conscious development (ISO 26262, IEC 61508), agentic workflows, AI-assisted development tools.
Responsibilities
Research and characterize real-world AV and physical AI workloads; Develop and optimize PVA algorithms; Translate workload insights into requirements for hardware, compilers, and SDKs; Build prototypes and performance models; Partner with teams to integrate PVA-accelerated algorithms into products; Publish technical findings.
Seniority
Senior, hands-on IC
