We are seeking a Senior Compiler Engineer to join a team of engineers focusedon Machine Learning, AI (including NPUs) and custom-built hardware solutions tied to security, speech and visual recognition, automated vehicles, and other leading edge industries. This individual will be a proficient C++ developer who will work with firmware, FPGA, and the model teams to optimize neural and digital signal processor hardware performance.
This is a contract to hire position located in San Diego. Relocation and H1B sponsorship is available.
Requirements & Responsibilities
- Develop compiler for that interfaces with deep learning frameworks such as Tensorflow, Google Cloud AI, Tensorflow, IBM Watson Analytics, Keras that converts neural networks into internal representations suitable for optimization.
- Develop optimization techniques and algorithms to map CNNs onto neural processors.
- Implement advanced code generation (source-to-source as well as binary).
- Develop supporting data compression techniques, algorithms, network pruning, etc.
- Create multiprocessor and multicore partitioning and scheduling strategies.
- Develop complex programs to validate the functionality and performance of the CNN application programming kit.
Requirements
- 3+ years of advanced compiler construction, target-independent optimization (using LLVM or other similar stacks).
- Extensive C++ and python experience in a production environment.
- Extensive experience with optimization, including: loop optimization, polyhedral models, and IR construction / transition / node lowering techniques.
- Familiarity with the latest deep learning compilation methodology is a huge plus.
- Knowledge of neural network exchange formats is a plus.