Pravi Devineni, Ph.D.

Senior Data Science Consultant

Duke Energy

Charlotte, NC

Interests: AI & Machine Learning,

NLP, Network Science
Data Mining, Anomaly Detection

  • LinkedIn
  • icons8-google-scholar-192
  • Twitter
  • Instagram

Contact me at
pravi (dot) valli (at) gmail (dot) com



Dr. Pravallika (Pravi) Devineni is a Senior Data Science Consultant at Duke Energy in Charlotte, NC, focused on solving problems related to electric grid. Prior to Duke Energy, Pravi was a Research Scientist and postdoctoral researcher in the Computer Science and Mathematics Division at Oak Ridge National Laboratory, Tennessee, USA. She received her Ph.D. from University of California Riverside in 2018, where her dissertation was about mining patterns and identifying changes in dynamic graph networks. Her interests include machine learning, graph networks, natural language processing, data science and its applications. 

Pravi has a passion for advocating for women in tech; she co-founded the Women in Computing groups at both UCR and UNM. She currently serves on the advisory board for GHC 2022 and was the chair for AI track at Grace Hopper Celebration of Women in Computing 2021. She can frequently be seen volunteering at women in tech events like Hour of Code.



2015 – 2018

University of California Riverside, CA

Ph.D. in Computer Science; Dissertation Title: "From Social Networks To Smartphones: Modeling And Understanding Online Human Behavior"

GPA: 3.84/4.0

2012 – 2015

University of New Mexico, Albuquerque, NM
Ph.D. in Computer Science (Moved to UCR)

2009 – 2011

Manipal University, Manipal, India

M.Tech in Computer Science and Engineering 

GPA: 8.57/10

Skills & Languages

Languages, Libraries and Tools
  • Python (including Pandas, scikit-learn, Matplotlib, Plotly, PyTorch, and Keras)

  • MATLAB, R, SQL, MongoDB, Neo4j

  • Git, SVN and Docker

  • HTML, CSS and Javascript

Machine Learning/Data Science Methods
  • Supervised Learning (logistic regression, random forest, gradient boosting, SVMs)
  • Unsupervised Learning (clustering, k-NN, anomaly detection, expectation maximization) 
  • Deep learning (convolutional neural networks, LSTMs, convolutional autoencoders, Word2Vec, Node2Vec)
  • Dimensionality reduction (SVD, NMF, PCA, tensor decomposition methods)
  • Graph Mining algorithms
  • Ensemble methods, unstructured data analysis, time series analysis

Work Experience

2022 – Present

Senior Data Science Consultant

Duke Energy

  • Developing ML models to predict dynamic trimming cycles for Vegetation Management

  • Problem solving to maintain Duke's electric grid infrastructure


Summer 2017

Computation Intern and Data Science Summer Institute (DSSI) Intern
Lawrence Livermore National Laboratory,
Livermore, CA

2020 – 2022

Research Scientist in Data Analytics
Oak Ridge National Laboratory

  • Design resource-efficient deep learning models

  • Help domain scientists find solutions

2018 – 2019

Postdoctoral Researcher, Computer Science and Mathematics Division

Oak Ridge National Laboratory, Oak Ridge, TN


Selected Publications

  1. Bill Kay, Hao Lu, Pravallika Devineni, Anika Tabassum, Supriya Chintavali, Sangkeun Lee, Identification of Critical Infrastructure via PageRank, IEEE BigData BTSD Workshop, 2021

  2. Uday Singh Saini, Pravallika Devineni, Evangelos E Papalexakis, Subspace Clustering Based Analysis of Neural Networks, ECML PKDD 2021

  3. Ravdeep S Pasricha, Pravallika Devineni, Evangelos E Papalexakis, Ramakrishnan Kannan, Tensorized Feature Spaces for Feature Explosion, ICPR 2021

  4. Pravallika Devineni, Bill Kay, Hao Lu, Anika Tabassum, Supriya Chintavali, Sangkeun Lee, Toward Quantifying Vulnerabilities in Critical Infrastructure Systems, IEEE BigData BTSD Workshop, 2020

  5. Steven R Young, Pravallika Devineni, Maryam Parsa, J Travis Johnston, Bill Kay, Robert M Patton, Catherine D Schuman, Derek C Rose, Thomas E Potok, Evolving Energy Efficient Convolutional Neural Networks,  IEEE BigData 2020

  6. Pravallika Devineni, Evangelos E. Papalexakis, Rama K. Vasudevan, Ramakrishnan Kannan, Convolutional Autoencoders for Unmixing High-dimensional Scientific Images, AI and Tensor Workshop, Santa Fe, NM, 2019

  7. Pravallika Devineni, Evangelos E. Papalexakis, Kalina J. Michalska, Michalis Faloutsos, MIMiS: Minimally Intrusive Mining of Smartphone User Behaviors, IEEE ASONAM, 2018

  8. Saba A Al-Sayouri, Pravallika Devineni, Sarah S Lam, Evangelos E Papalexakis, Danai Koutra, GECS: Graph Embedding Using Connection Subgraphs, SIGKDD MLG Workshop 2017

  9. Pravallika Devineni, Evangelos E Papalexakis, Danai Koutra, A Seza Dogruöz, Michalis Faloutsos, One Size Does Not Fit All: Profiling Personalized Time-Evolving User Behaviors, IEEE ASONAM 2017

  10. Pravallika Devineni, Danai Koutra, Michalis Faloutsos, Christos Faloutsos, If Walls Could Talk: Patterns and Anomalies in Facebook Wallposts, IEEE ASONAM 2015

Professional Service & Outreach 

Program Committee Member 

  • Pacific-Asia Conference on Knowledge Discovery and Data Mining

  • SIGKDD Conference on Knowledge Discovery and Data Mining

  • Grace Hopper Conference

Journal Reviewer

  • IEEE Transactions on Mobile Computing (TMC)

  • IEEE Transactions on Neural Networks and Learning Systems (TNNLS)

  • Springer Data Mining and Knowledge Discovery (DAMI)

Workshop Organization

  • Chair, SMC Data Challenge 2021

  • Co-organizer, IEEE BigData BTSD Workshop 2021

  • Co-chair - GHC 2021 AI Track 


  • "Surviving a Ph.D.", SIAM SDM Doctoral Forum 2021


  • Thomas Reichel, Undergraduate at UIUC, Summer 2021

  • JaCoya Thompson, PhD Student at Northwestern University,  GEM Fellow, Summer 2021

  • Ravdeep Pasricha, PhD Candidate at University of California Riverside, Summer 2019