Role overview
We are looking for an Engineer / Research Scientist with deep expertise in quantized deep learning models for hardware acceleration in autonomous systems. In this cross-disciplinary role, you will bridge perception model design and hardware-aware deployment, enabling efficient execution of high-performance perception algorithms across embedded compute platforms. You will focus on researching state of the art perception models and develop optimization pipelines for the quantized versions of these models customized to provide real-time performance and energy efficiency on next-generation autonomy hardware.
- Research state of the art perception models in collaboration with the ADAS SW teams
- Lead the development of optimizations for mapping quantized perception models (e.g., CNNs, Transformers, LLMs) to embedded and heterogeneous hardware platforms.
- Design and implement hardware-aware optimizations, including quantization strategies, model compression, memory-efficient representations, and operator fusion, targeted to custom accelerators.
- Collaborate with hardware teams to co-optimize model architecture and compute pipeline under real-time constraints (latency, throughput, power).
- Benchmark and analyze system performance across platforms and iterate to achieve optimal deployment efficiency.
Partner with perception, systems, and autonomy teams to align model optimization efforts with hardware roadmap and real-world autonomy requirements.
Ph.D. or M.S. in Computer Engineering, Electrical Engineering, Computer Science, or related field with a focus on ML compilers, embedded systems, or hardware-aware AI.
Hands-on experience with quantized model deployment, ML design stacks, and code generation for embedded or heterogeneous compute systems.
Strong understanding of computer vision models (e.g., object detection, segmentation) and their optimization for edge inference.
Proficiency in deep learning frameworks (e.g., PyTorch, TensorFlow) and their low-level IRs or export formats (e.g., ONNX).
Solid programming skills in C++, Python
Familiarity with CUDA/OpenCL (or other accelerator programming models).
What we're looking for
- Prior experience working with hardware-software co-design, especially for autonomous or robotics platforms.
- Deep knowledge of numerical precision trade-offs, quantization-aware training (QAT), and dynamic/static quantization flows.
- Familiarity with embedded real-time constraints and hardware profiling/debugging tools.
- Familiarity with rearchitecting models to best suit hardware capabilities.
- Publication record in top-tier ML/Systems conferences (e.g., MLSys, NeurIPS, DAC, ICCAD).