Senior Compiler Engineer

Job description


  1. Develop a deep learning compiler stack that interfaces frameworks such as Tensorflow, Caffe2Keras etc. and converts neural nets (CNN/RNN) into internal representations suitable for optimizations. 
  2. Develop new optimization techniques and algorithms to efficiently map CNNs onto Kneron NPU processors  
  3. Implement state of the art code generation (source-to-source as well as binary) 
  4. Develop supporting data compression techniques, quantization algorithms, tensor sparsity enhancements, network pruning, etc 
  5. Devise multiprocessor/multicore partitioning and scheduling strategies 
  6. Develop complex programs to validate the functionality and performance of the CNN application programming kit 
  7. Performance analysis of kernels, benchmarks, and CNN applications. 
  8. New compiler feature development and debugging. 
  9. Help in authoring and reviewing product documentation 
  10. Assist application engineering team support customers of the product (some amount of direct customer interaction may be required). 

Requirements

  1. 3-5+ years of experience working on a production compiler.  
  2. Advanced compiler construction, target-independent optimizations and analyses, code generation fundamentals is a must. 
  3. Expertise in software development, test, debug and release required. 
  4. Great C++ is a must, Python mandatory, but less pressing. 
  5. Knowledge of and experience with LLVM compiler stack is very desirable (other state-of-the-art compilers qualify too). 
  6. High to intermediate optimization space: loop optimization, polyhedral models, IR construction/transition/lowering techniques is a big plus. 
  7. Prior work with CNNs and familiarity with deep learning frameworks (Tensorflow, Caffe/2, etc.) is a strong plus. 
  8. Familiarity with the state-of-the-art deep learning compilation approaches is a huge plus: XLA, Glow, ONNX, Tensor Comprehensions, etc. 
  9. Familiarity with various deep learning networks and their applications (Classification/Segmentation/Object Detection/RNNs) is a plus. 
  10. Knowledge of neural net exchange formats (NNVM, NNEF) is a bonus 

Location

10052 Mesa Ridge Court, Suite 101 San Diego, CA 92121

Contact

If interested, please send your resume to:  career@kneron.us