The Machine Learning Infrastructure team builds the industrial scale machine learning platform used by our Machine Learning engineers developing software for our self-driving car. We provide essential tools and frameworks to support the entire lifecycle of machine learning, from data processing, large scale training and evaluation frameworks to efficient neural net inference runtimes for onboard execution and simulation. As the team lead, you will oversee the team responsible for defining, defining and implementing the compute platform.
The Motional global headquarters are located at 100 Northern Avenue in Boston, MA. Nestled in the bustling Seaport district with sweeping views of Boston Harbor and downtown Boston, the offices are located close to major transit lines and a quick walk to various restaurants and popular attractions.
What You'll Be Doing
- Manage and monitor ML training, visualize and compare ML training, evaluate models on different hardware accelerators and track cost, etc.
- Own training cluster and provide technical support to internal customers.
- Inspire and motivate the team to work together as a cohesive and productive unit
- Work closely with machine learning researchers to identify the bottlenecks and optimize the codebase by: better parallelism and data caching, fully utilizing the hardware: CPU, GPU, memory and disk/network IO, distributed/multi-machine training.
- Act as a sounding board and technical architect for your teams, help define technical vision and strategy
- Provide coding and performance guidelines and best practices to machine learning researchers/engineers.
What We're Looking For
- Degree in Computer Engineering, Computer Science, Applied Mathematics, or a related field
- 5+ years of industry experience, including 3+ years as Tech Lead or Manager
- Strong programming skills in Python.
- Familiar with DL frameworks like PyTorch, Tensorflow, Keras, etc.
- Experience working with cloud infrastructure like AWS, GCP or Azure.
- Experience working with large scale dataset.
- Strong skills in C++.
- Experience in numba, cython, CUDA, etc.
- Experience in distributed computing.
- Strong machine learning/deep learning background.
Why you should join us:
- Youll have the opportunity to work on cutting-edge technology and some of the most exciting problems within engineering including: robotics, infrastructure, visualization, etc.
- As part of the Machine Learning Infrastructure team, youll be collaborating and working closely with a tight knit team and some of the industrys top researchers.