As part of OPENCHIP & SOFTWARE TECHNOLOGIES Software Design group, you will design and manufacture our next-generation, high-performance, power-efficient processor, system-on-chip (SoC) AI accelerators software environment. The target product is a series of AI-specific semiconductor systems designed to deliver low power consumption and high-performance chips providing efficiency in machine learning and deep learning applications. These chips are used in various industries, including HPC, security, manufacturing, mobility, automotive, and more.
·As Software AI/ML Compiler engineer you will drive the development of the AI accelerator compiler. You will be responsible of writing some of the key components of the software involved in mapping AI/ML models from frameworks such as PyTorch, TensorFlow, TVM, MLIR to the AI accelerator system architecture. You are someone that has insight into performance optimizations and will be able to think of novel solutions, implementing them from concept to production.
- Experience in compiler frameworks such as TVM, MLIR, LLVM.
- Expertise with Deep Neural network architectures like CNN, RNN.
- Proven record of enabling AI accelerator using AI frameworks like TensorFlow, PyTorch, ONNXRT, TVM, MLIR.
- Understanding of Computer Architecture and data Arithmetic with a focus on RISC V processor’s architecture.
- Understanding of AI application stacks as LLM, ASR, Generative AI.
- Understanding of the flow of AI model optimization techniques like quantization, sparsity, etc.
- Knowledge of Acceleration platforms like GPU, TPU, APU, FPGAs.
- Experience in Software Development delivering C++ applications