Chief Engineer: Deep Learning Compiler Expert

Job Title: Chief Engineer: Deep Learning Compiler Expert

You will collaborate with experts in machine learning, algorithms and software to lead our effort of deploying machine learning models onto Samsung Mobile AI platform.

In this position, you will contribute, develop and enhance our compiler infrastructure for high-performance by using open-source technology like MLIR, LLVM, TVM and IREE.

Necessary Skills / Attributes:

• 6 to 15 years of experience in the field of compiler design and graph mapping.
• 2+ years hands-on experience with MLIR and/or LLVM.
• Experience with multiple toolchains, compilers, and Instruction Set Architectures.
• Strong knowledge of resource management, scheduling, code generation, and compute graph optimization.
• Strong expertise in writing modern standards (C++17 or newer) C++ production quality code along test-driven development principles.
• Comfortable and experienced in software development life cycle - coding, debugging, optimization, testing, and continuous integration.
• Familiarity with parallelization techniques for ML acceleration.
• Experience working on and contributing to an active compiler toolchain codebase, such as LLVM, MLIR, or Glow.
• Experience in deep learning algorithms and techniques, e.g., convolutional neural networks, recurrent networks, etc.
• Experience of developing in a mainstream machine-learning framework, e.g. PyTorch, Tensorflow or Caffe.
• Experience operating in a fast-moving environment where the workloads evolve at a rapid pace.
• Understanding of the interplay of hardware and software architectures on future algorithms, programming models and applications.
• Experience developing innovative architectures to extend the state of the art in DL performance and efficiency.
• Experience with Hardware and Software Co-design.

M.S. or higher degree, in CS/CE/EE or equivalent with industry or open-source experience.

Work Profile:

• Design, implement and test compiler features and capabilities related to infrastructure and compiler passes.
• Ingest CNN graphs in Pytorch/TF/TFLite/ONNX format and map them to hardware implementations, model data-flows, create resource utilization cost-benefit analysis and estimate silicon performance.
• Develop graph compiler optimizations (operator fusion, layout optimization, etc) that are customized to each of the different ML accelerators in the system.
• Integrate open-source and vendor compiler technology into Samsung ML internal compiler infrastructure.
• Collaborate with Samsung ML acceleration platform engineers to guide the direction of inferencing and provide requirements and feature requests for hardware vendors.
• Closely follow industry and academic developments in the ML compiler domain and provide performance guidelines and standard methodologies for other ML engineers.
• Create and optimize compiler backend to leverage the full hardware potential, efficiently optimizing them using novel approaches.
• Evaluate code performance, debug, diagnose and drive resolution of compiler and cross-disciplinary system issues.
• Contribute to the development of machine-learning libraries, intermediate representations, export formats and analysis tools.
• Communicate and collaborate effectively with cross-functional hardware and software engineering teams.
• Champion engineering and operational excellence, establishing metrics and processes for regular assessment and improvement.

Keywords to source candidates

Senior Developer, Deep Learning, Prediction engine, Machine Learning, Compiler