Requisition ID # 162168 

Job Category: Accounting / Finance; Information Technology 

Job Level: Individual Contributor

Business Unit: Information Technology

Work Type: Hybrid

Job Location: Oakland

 

 

Department Overview

The Data Science & Artificial Intelligence Department consists of a “Delivery” team that develop data science and machine learning solutions.

As a Delivery team, this Department uses industry leading data science and change management practices to drive PG&E’s transition to the sustainable grid of the future. The Department works cross-functionally across the company to enable data driven decisions applying analytics, as well as improvements to relevant business processes. Deployed to some of PG&E’s highest priority arenas, the Department does not specialize in a traditional utility domain, such as asset management or program administration, but instead specializes in extracting useful insights from disparate datasets and facilitating actions informed by these insights.

 

The team works on a wide variety of complex business problems, offering constant opportunity to explore and learn. Current and past engagements include:

· Creating wildfire risk models that are used by regulators and the utility to prioritize asset management

· Developing computer vision models that improve, accelerate, and automate asset inspections processes

· Predicting electric distribution equipment failure before it occurs, allowing for proactive maintenance

· Forming the analytical framework behind PG&E’s Transmission Public Safety Power Shutoff

· Optimizing non-wires alternative resource portfolios, like the Oakland Clean Energy Initiative, including location and resource adequacy considerations

· Analyzing customer demographic, program participation, and SmartMeter interval data to build program targeted propensity models, e.g. for customer owned distributed energy resource technologies like EVs

· Identifying and investigating anomalous customer natural gas usage, to resolve dangerous customer side gas leaks

 

Position Summary

PG&E is looking for a Data Scientist with experience in delivering data science products end-to-end. In this role, the successful candidates will be uniquely positioned at the forefront of utility industry analytics, having the opportunity to advance PG&E’s triple bottom line of People, Planet, and Prosperity. Working as part of cross functional teams, including data engineers, data scientists, technologists, subject matter experts, and change management professionals – this individual will lead the development of machine learning models to identify electric vehicles charging loads from AMI data disaggregation (Advance Metering Infrastructure). The data scientist in this role will also help build machine learning models to predict the next customers who will adopt an EV based on their socioeconomics, energy consumption attributes and other characteristics. The individual will participate in the full lifecycle of the delivery process from initial value discovery to model-building to building data products to deliver value to end users.

 

This position is hybrid, working from your remote office and the Oakland General Office(OGO) at least once per week and or as business needs require.

 

PG&E is providing the salary range that the company in good faith believes it might pay for this position at the time of the job posting. This compensation range is specific to the locality of the job. The actual salary paid to an individual will be based on multiple factors, including, but not limited to, specific skills, education, licenses or certifications, experience, market value, geographic location, and internal equity. Although we estimate the successful candidate hired into this role will be placed between the entry point and the middle of the range, the decision will be made on a case-by-case basis related to these factors.​ This job is also eligible to participate in PG&E’s discretionary incentive compensation programs.  

 

A reasonable salary range is:​

 

Bay Area Minimum:     $122,000  

Bay Area Maximum:    $194,000  

 

 

Job Responsibilities

 

Design and develop production-quality scientific algorithms in Python to extract patterns of customer energy consumption, as well as other customers’ characteristics and attributes. Develop and apply advanced statistical and machine learning models to disaggregate energy load data from coarse-grained sources such as Advanced Metering Infrastructure, Customer Billing, Customer Energy Programs, etc. Develop spatio-temporal algorithms to predict adoption of Electric Vehicles by customers. Perform in-depth validation of our algorithms that is driven by business and technical requirements. Perform deep root-cause analysis, EDA, and error analysis of the ML models. Collaborate with members of your team and with domain experts (e.g., Power Distribution, Grid Planning, Clean Energy Transportation) to understand practical implications of your solution, to collect business requirements and to deliver results to business partners. Communicate technical information, their implications and applications to peers, various business partners, and strategic leaders across the company.

 

 

Qualifications

Minimum:

 

Bachelor’s degree +4 years or MS +2 yrs of experience, in Engineering, Computer Science, Physics, Econometrics or Economics, Mathematics, Applied Sciences, Statistics, or other highly quantitative discipline. Demonstrated knowledge of and abilities with data science standards and processes (model building and evaluation, optimization, feature engineering, etc.) along with best practices to implement them. Experience with handling large datasets and cloud computing platforms (e.g. AWS, Azure, GCP, or other enterprise level analytics platforms. Proficiency in programming languages such as Python and/or R. Experience designing, developing, and maintaining scientific code that runs at scale. Strong understanding of applied statistics and probability. Experience with spatio-temporal statistics, geospatial data analysis, and machine learning techniques for time series with large datasets.

 

Desired:

Ph.D. in Engineering, Computer Science, Physics, Econometrics or Economics, Mathematics, Applied Sciences, Statistics, or other highly quantitative discipline. Relevant industry experience (electric or gas utility, EV charging infrastructure, distributed energy resources, analytics consulting, etc). 2+ years of experience writing software to extract features from time series data or large-scale datasets. Experience turning business needs into technical requirements, and structuring developments, analysis and validation plans to meet requirements. Excellent analytical, problem solving, and communication skills.

#featuredjob