Akshay Kale

Artificial Intelligence | Data Science | Research

Résumé CV

Summary

5+ years of experience in machine learning modeling with publications
Academic research focused on explainable artificial intelligence
Ph.D. in Artificial Intelligence, completed May 2024

Experience→

Data Scientist – Kiewit Corporation

April 2024 – Current

Built machine learning models for safety risk forecasting, productivity prediction, and cost-risk analysis
Supported analytics across scheduling, estimating, indirect cost benchmarking, and craft overrun modeling
Communicated results through dashboards, reports, and executive-level stakeholder presentations

Machine Learning Engineer – University of Nebraska (UNO)

January 2017 – May 2024

Trained machine learning models with 98% accuracy in structural health monitoring
Developed deep learning models for bridge maintenance prediction, resulting in 3 research publications
Built data pipelines collecting over 21 million inspection records

Selected Projects→

A LSTM Deep Learning Model for Predicting and Recreating Sequences of Missing Data Big idea — Classical machine learning models are designed to work with fixed length inputs. Small idea — Learning the observation's temporal ordering can make it challenging to extract features suitable for use as…

GitHub ↗ Open full page →

↗

CoRA

Data Science | D3.js | JavaScript

Developed an interactive Forced Directed Network Graph visualization to identify fallen World War II soldiers from their remains

Objective Build an intelligent Tic-Tac-Toe player using Q-Learning — watch as the program learns to outsmart you, one game at a time. Getting Started Clone the repository Run the game

GitHub ↗ Open full page →

Top Three Publications→

↗

Ramsey A, Kale A, Kassa Y, Ricks B, Gandhi R. Toward Interactive Visualizations for Explaining Machine Learning Models. Proceedings of the Information Systems for Crisis Response and Management Conference, Omaha, NE, USA. 2023.

↗

Kale A, Kassa Y, Ricks B, Gandhi R. A Comparative Assessment of Bridge Deck Wearing Surfaces: Performance, Deterioration, and Maintenance. Applied Sciences. 2023; 13(19):10883.

↗

Kale A, Ricks B, and Gandhi R. New measure to understand and compare bridge conditions based on inspections time-series data. Journal of Infrastructure Systems 27.4 (2021): 04021037.