Machine Learning - Architect/Researcher/Engineer

Machine Learning - Architect/Researcher/Engineer - Accelerator, Deployment and Infrastructure

[Overview]
MediaTek HQ is looking for Machine Learning Architect, Researcher, or Engineer who is passionate about System Platform, Software Engineering, and Applied Machine Learning (including Deep Learning and AI). This position is within MediaTek’s AI Group.

This R&D role offers a dynamic environment, hands-on learning, and a unique opportunity to develop both Infrastructures and Accelerators for state-of-the-art (SOTA) machine learning, as well as to build applications utilizing SOTA machine learning techniques.

You will join a team devoted to developing and deploying cutting-edge machine learning and deep learning algorithms, optimized for MediaTek’s NPU—and, conversely, optimizing our NPU for advanced ML/DL workloads.

#LI-AMY

[Responsibility]
As a ML Architect, the major responsibility is to derive and define next generation NPU capability/architecture in an Application x HW x SW style. The other responsibility also includes learning and adopting relevant new technologies, framework, packages, processes and tools for NPU architecture.
As a ML Researcher, the major responsibility is to conduct exciting research/development, present findings and transform these into practical technology. The other responsibility also includes learning and adopting relevant new technologies, framework, packages, processes and tools in the research.
As an ML Engineer, the major responsibility is to solve problems and build innovative software solutions. The other responsibility also includes learning and adopting relevant new technologies, framework, packages, processes and tools in the proposed solutions.
[Additional Responsibility]
+ Contribute ideas and publish high impact papers for the project that includes findings and results.
+ Create research prototypes exploring AI and its applications.
+ Contribute to the team by Accelerator, Deployment and/or Infrastructure
+ Make development impact by creating architectural changes if Accelerator
+ Make development impact by creating prototype if Deployment
+ Make development impact by creating solution (including SDK, API, … and documentation) if Infrastructure
[Required Qualifications]
+ Master’s degree or higher in computer science, electrical engineering or related fields.
+ 2+ year experience in machine learning, deep learning, computer vision, signal processing, other ML topics.
+ 2+ year coding experience in python, C++, JAVA (at least one of).
+ Prior hands-on experience in ML frameworks (like Keras, Tensorflow, PyTorch, or …).
[Preferred Qualifications]
+ Ph.D. candidate with publication experience in top conferences/journals, like ISSCC, VLSI, ISCA, MICRO, HPCA, DAC, PACT, PPOPP, CVPR, ICCV, ECCV, ICML, NeurIPS, IJCAI, AAAI, etc.
+ Demonstrated understanding of data structures, algorithms and/or other computer & software architecture factors that affect code quality, performance and user experience.
+ Experience creating research prototypes and the implementation of end-to-end machine learning pipeline.
+ Experience building prototype/software for hackathon, research project and so on.
+ Ability to conduct original and novel research.
+ Analytical and problem-solving skills.
+ Ability to both work alone and collaborate effectively with other team members.
+ Experience documenting findings, results and software documentation.