Director – Machine Learning Compiler Development

San Francisco Bay Area, USA

We are seeking an experienced software engineering professional to lead our Compiler and Toolchain team: design, implement and verify the tools that underpin our ultra-low-power Edge AI solutions.

Responsibilities

  • Lead development of our deep learning compiler stack that maps the output of frameworks such as Tensorflow, PyTorch, Keras onto power-efficient edge AI hardware
  • Develop new optimization techniques and algorithms
  • Devise multiprocessor/multicore partitioning and scheduling strategies
  • Manage performance trade-offs; understanding the balance between performance, memory, and power in compiler generated code

Qualifications

  • Demonstrated ability to deliver robust, high-reliability software products for technical users
  • 5+ years leadership experience
  • Proficiency in C/C++
  • Excellent theoretical and practical understanding of chip-level hardware architectures – memory bandwidth, latency, caches, vector processing, IPC etc
  • Experience with application-focused hardware acceleration technologies, such as GPU acceleration with CUDA or OpenCL, or FPGA acceleration with OpenCL or CAPI
  • Strong understanding of compiler technology stacks
  • Strong CS fundamentals:  object-oriented design, data structures and algorithm design, complexity analysis, scalability, and availability
  • Agile development experience
  • Excellent verbal and written communication skills

Preferred

  • Machine Learning experience
  • Tensorflow/Tflite experience
  • Familiarity with computer vision algorithms such as object detection, tracking, and recognition
  • Prior work with CNNs and familiarity with deep learning frameworks (Tensorflow, PyTorch, etc.)
  • Familiarity with the state-of-the-art deep learning compilation approaches: XLA, Glow, ONNX, Tensor Comprehensions
  • Experience with optimization techniques for neural networks such as quantization, pruning, and distillation
APPLY TO THIS JOB