The Energy and Water team is looking for a senior/principal level data scientist to join our data science organization. Utilities from around the world provide energy usage data from tens of millions of customers at some of the largest utilities.
We work across a breadth of areas, from customer-facing aspects of the energy and water business to generation, transmission and distribution on the grid side. We use data science to solve real-world problems like helping make energy bills more affordable, keeping the lights on after storms, and helping utilities transition to renewable sources of energy.
Career Level – IC4
Responsibilities include a mix of the following tasks:
- Stay on top of current trends in data science and find avenues to apply those tools to problems in our backlog.
- Our data science backlog (in the area of Distributed Energy Management and Outage Management, for example) are critical path for product. Candidates for this role/level will be expected to own delivery of productized solutions. This will involve engaging in early-stage discussions with PMs and client-facing teams to help guide them towards writing requirements that enable feasible data science models to be deployed across clients, and work matrixed (or as a tech lead) to other data scientists to build out the solution to the posed problem.
- Prior experience in building/deploying GenAI solutions (for example, using available foundational models, or RAG training on top of these models), and/or Computer Vision solutions is a plus.
- Our data science backlog (in the area of Distributed Energy Management and Outage Management, for example) are critical path for product. Candidates for this role/level will be expected to own delivery of productized solutions. This will involve engaging in early-stage discussions with PMs and client-facing teams to help guide them towards writing requirements that enable feasible data science models to be deployed across clients, and work matrixed (or as a tech lead) to other data scientists to build out the solution to the posed problem.
- Identify recurring problems and bottlenecks in our data science R&D stack and develop/recommend solutions.
- This will require experience and understanding of modern data science development and production stacks, preferably in the OCI context, but experience in other similar stacks will also be considered.
- Assist manager and leaders in refining scope of prioritized tasks by identifying required datasets, building proofs-of-concept models, identifying gaps.
- Efficiently execute defined, data science tasks, as prioritized by leadership.
- Productize built models using our OCI data science stack
- Work with our client management team and our customers to define and take ownership of scope, intermediate deliverables, and timelines around data science deliverables.
- Mentor junior data scientists and recommend best research practices for the team;
Preferred Qualifications
- PhD in Computer Science, Mathematics, Statistics, Physics, Economics or other STEM field; strong MS/BS candidates will also be considered. Prior energy sector experience is a plus but not a gating requirement.
- 8-10 years of professional work experience in a data science role spanning either pure data science research or a mix of model building and MLOps. PhD candidates with lower tenure in senior roles will also be considered.
- Prior experience with Python, PyTorch (or other equivalent deep learning frameworks) strongly preferred. PySpark nice to have.
Strong written & oral communication skills;
A successful candidate will demonstrate the following qualities:
- Prior experience working as a data scientist (either research, R&D or MLOps)
- Prior experience working in cloud data science environments to build and validate ML models.
- Breadth and depth across machine learning techniques: specifically Deep Learning.
- A reasonable number of our problems also benefit from classical ML techniques, so prior experience crafting features to train models is also required.
APPLY
- A reasonable number of our problems also benefit from classical ML techniques, so prior experience crafting features to train models is also required.