Senior Software Engineer - Applied Research in Robotics

NVIDIA NVIDIA · Semiconductors · Bangalore, India

Senior Software Engineer focused on applied research in robotics, optimizing robotics solutions on NVIDIA's Programmable Vision Accelerator (PVA). The role involves mapping high-impact robotics workloads to efficient execution on heterogeneous hardware, improving latency, determinism, and power efficiency for real-time systems. Responsibilities include designing and optimizing PVA operators, researching parallelization strategies, and collaborating with various teams. The role also leverages AI-assisted development tools and agentic workflows.

What you'd actually do

  1. Research and analyze end-to-end robotics software stacks to discover, characterize, and prioritize high-impact workloads that can be accelerated on PVA.
  2. Design, implement, test, and optimize PVA operators for robotics workloads, including end-to-end integration into real robotics pipelines.
  3. Research parallelization and data movement strategies for mapping robotics algorithms onto PVA efficiently!
  4. Work across camera, RADAR, LiDAR, IMU, perception, planning, and actuator paths to understand algorithmic requirements, memory behavior, timing constraints, and real-time execution needs.
  5. Build benchmarks, validation infrastructure, and technical mentorship for robotics teams adopting PVA acceleration!

Skills

Required

  • BS/MS or PhD in Computer Science, Robotics, Electrical Engineering, Computer Engineering, or equivalent experience.
  • 5+ years of experience in robotics software, embedded systems, accelerated computing, computer vision, sensor processing, perception systems, or related areas.
  • Strong robotics algorithm knowledge, with practical understanding of perception, sensor fusion, control, planning, or actuator pipelines.
  • Strong C/C++ programming skills, including low-level debugging, performance profiling, optimization, and production-quality software development.
  • Experience analyzing algorithms and mapping workloads to heterogeneous hardware such as CPUs, GPUs, DSPs, SIMD/VLIW processors, or domain-specific accelerators.
  • Deep understanding of embedded systems, SoC architecture, memory hierarchy, cache behavior, DMA, real-time scheduling, and performance bottlenecks.
  • Experience with embedded Linux and/or QNX development environments.
  • Experience with robotics middleware, ROS 2, real robot bring-up, sensor integration, or actuator/control paths.

Nice to have

  • Background with NVIDIA Jetson, DRIVE, IGX, ISSAC, VPI, NvMedia, or PVA.
  • Experience building robotics applications that use CUDA or other accelerators for perception, planning, simulation, or real-time sensor processing.
  • Familiarity with safety-conscious development processes such as ISO 26262, IEC 61508, or similar standards.

What the JD emphasized

  • 5+ years of experience in robotics software, embedded systems, accelerated computing, computer vision, sensor processing, perception systems, or related areas.
  • Strong robotics algorithm knowledge, with practical understanding of perception, sensor fusion, control, planning, or actuator pipelines.
  • Strong C/C++ programming skills, including low-level debugging, performance profiling, optimization, and production-quality software development.
  • Experience analyzing algorithms and mapping workloads to heterogeneous hardware such as CPUs, GPUs, DSPs, SIMD/VLIW processors, or domain-specific accelerators.
  • Deep understanding of embedded systems, SoC architecture, memory hierarchy, cache behavior, DMA, real-time scheduling, and performance bottlenecks.
  • Experience with robotics middleware, ROS 2, real robot bring-up, sensor integration, or actuator/control paths.

Other signals

  • robotics acceleration
  • PVA
  • heterogeneous computing
  • real-time systems
  • AI-assisted development tools
  • agentic workflows