Mara Legner
Principal Data Scientist | Machine Learning, Deep Learning, NLP, LLMs, Generative AI
- Role
- Principal Data Scientist at Duke Energy Corporation
- Location
- Chicago, IL, US
- LinkedIn followers
- 500 followers
About Mara Legner
I\'m a problem solver at heart. In the past few years I\'ve taken on a lot of home renovation work which means doing a lot of pre-planning and research, making a ton of decisions on sizes/finishes/colors, dealing with unforeseen issues on the fly, and realizing that nothing is ever level, square, or plumb. I also enjoy traveling and a big component is figuring out logistics of where to go, how to get there, where to stay, things to do, and what to pack. I also love all types of puzzles: jigsaw, crossword, Wordle, Sudoku (3 time UW Sudoku Master champion).Throughout my career I\'ve applied this problem solving ability to a wide variety of industries including energy, utilities, aviation, and ratings and reviews. I enjoy researching models and statistical approaches, data wrangling and cleaning, testing and training models, putting models into production, tracking model performance, and iterating as needed. I love solving problems with data but data has limitations, so I also place high value on business stakeholder and SME input, non-quantitative metrics, and keeping humans in the loop.I have experience with all of the following:Large Language Models (LLMs), Generative AI, Deep Learning, Neural Networks, Natural Language Processing (NLP), Machine Learning, Sentiment Analysis, Regular Expressions (RegEx), Python, SQL, PySpark, Amazon Web Services (AWS), Google Cloud Platform (GCP), Hugging Face, H2O Studio, Snowflake, dbt, Github, Tableau, Time Series (ARIMA), SAS
Experience
Principal Data Scientist
Apr 2025 — Present
Built and optimized 41 machine learning models (linear regression, XGBoost) for feeder-level energy forecasting, each achieving <5% SMAPE• Detected outliers in energy usage by applying clustering methods (DBSCAN, Isolation Forest, OPTICS, LOF) • Coded and productionalized SMAPE and MAPE calculations at the single and three phase feeder levels across the entire service area
Education
University of Wisconsin-Madison
Bachelor of Science (BS), Chemical and Biological Engineering
2006 — 2010
UCLA
Master of Financial Engineering
2013 — 2014
Skills
- Analysis
- R
- Chemical Engineering
- Management
- Technical Writing
- Data Analysis
- Python
- Financial Modeling
- Matlab
- Sas Programming
- Process Simulation
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.