5 hours ago
Milpitas, CA, USASenior / Staff+
H1B sponsor
Base Salary
$185k - $325k/yr
Responsibilities
- Develop and optimize real-time inference pipelines for embodied AI models on edge hardware.
- Implement custom CUDA kernels, memory-layout tuning, and hardware-aware graph compilation.
- Apply quantization, pruning, distillation, and structured sparsity for production-grade inference performance.
- Profile and debug inference stacks using Nsight, TensorRT, and Triton.
- Collaborate with ML research and robotics teams to meet real-time control-loop latency requirements.
- Create benchmarking frameworks covering latency, throughput, power consumption, and accuracy.
Requirements
- Master’s degree in Computer Science, Electrical Engineering, or a related field with 4+ years of experience, or a PhD.
- Deep expertise in CUDA programming, GPU architecture, low-level kernel optimization, and custom kernel authoring.
- Hands-on experience with quantization, pruning, distillation, and deployment using TensorRT, ONNX Runtime, TVM, or Triton.
- Proficiency in C++ and Python with strong systems programming and performance profiling skills.
- Experience deploying ML models on edge or embedded hardware such as NVIDIA Jetson, Orin, or equivalent ARM/GPU SoCs.
- Bonus: familiarity with VLA, multimodal transformers, diffusion-based policies, XLA, torch.compile, MLIR, robotics timing constraints, or efficient deep learning and on-device ML publications or production systems.
Benefits
- Competitive stock options and equity programs.
- Health, dental, and vision insurance plus a 401(k) plan.
- Visa sponsorship and green card support for qualified candidates.
- Lunches and dinners, a fully stocked kitchen, and regular team-building events.
- Full-time U.S. position requiring five days per week in-office collaboration.
Categories
AI ResearchML Engineering
About RoboForce
RoboForce builds AI-driven robotic systems designed to perform dull, dirty, and dangerous tasks in demanding industrial environments. Its products focus on deployable robo-labor that automates physical work for industrial operators, with an emphasis on real-world scalability and reliability. The privately held company was founded in 2023 and is headquartered in Milpitas, California.