I am a senior data scientist at Illumination Works, developing AI, ML and Data Science applications for the US Air Force and the Air Force Research Laboratory. Prior to joining ILW, I was on the faculty of Manhattan University and Caltech, a Fulbright-Nehru Visiting Scholar/Professor at the Indian Institute of Technology Madras, and on the building engineering team at the international multi-disciplinary firm Arup. My current interests and expertise span the data science areas of time series forecasting, anomaly detection, natural language processing and large language models, and computer vision. In recent times, I have also worked on damage detection in buildings using machine learning algorithms trained on large synthetically generated datasets from digital-twin simulations. I use the Python stack and Keras/Tensorflow extensively. On the engineering front, I am licensed as a Professional Civil Engineer (PE) and a Structural Engineer (SE) in the State of California. I serve on the Editorial Board of ASCE's flagship journal, the Journal of Structural Engineering, as an Associate Editor.
News:
4/9/2025: My two-part blog on uncertainty in AI applications is online. Part 1: "Uncertainty: What is it and why should you care?" [Click here to read Part 1]. Part 2: "Uncertainty in AI domains" [Click here to read Part 2].
2/5/2025: The paper titled "Retrofit Strategies Using Braces for Pre-Northridge Steel Moment-Frame Buildings: An Analytical Study", based on my student Arnar Bjorn Bjornsson's PhD research, has been published in the Journal of Earthquake Engineering [doi]. [PDF].
12/1/2024: Three time series forecasting applications, an ARIMA-on-demand (univariate) application, a SARIMA-on-demand (univariate) application, and a multi-variate deep learning-based application [including auto-regressive (AR), dense neural network (DNN), Long Short-Term Memory (LSTM), deep LSTM, and bidirectional LSTM models using the TensorFlow ML platform and the Keras API on Databricks] that I developed for the US Air Force LCMC have been deployed to production. Read more on my LinkedIn post: [LinkedIn].
7/22/2022: My paper titled "A Collapse Mechanics-Based Criterion for the Optimal Proportioning of Steel Moment Frames Subjected to Earthquakes" has been published in the ASCE Journal of Structural Engineering [doi]. [PDF].
Watch the NOVA episode "High-Risk High-Rise" online or on your local PBS channel, where I talk about the risk posed to tall buildings from strong earthquakes.
12/17/2021: My Manhattan College colleagues Professors Juneseok Lee (PI), Matthew Volovski, Mehdi Omidvar, Christina Cercone, Medya Fathi, Peter Sweeney, and I have been awarded a 3-year grant by the US National Science Foundation for our proposal titled, "Developing Students' Systems Thinking and Data Analytics Skills in Civil and Environmental Engineering". Award Number: 2142131. Project Duration: 05/2022-10/2024. Grant Amount: $399K.
Python code for ingesting files into the MySQL rupture-to-rafters simulations database is available for download here: https://github.com/swamikrishnan/Rupture-to-Rafters-Simulation-MySQL-Database.
Announcing the release of V1.0 of a MySQL database of rupture-to-rafters simulations. Get in touch for access.
FRAME3D, our three-dimensional nonlinear analysis software for buildings and other structures is available for download on GitHub: https://github.com/swamikrishnan/FRAME3D.
FRAME3D GUI is available for download on GitHub: https://github.com/swamikrishnan/FRAME3D-GUI.
Catch my presentation (Abstract 2031600) titled, "Performance Quantification of Tall Steel Braced Frame Buildings Using Rupture-To-Rafters Simulations", at the EMI 2021 Virtual Conference, May 25-28, 2021. Click here to watch the video.
Catch my presentation (Abstract 2031644) titled, "An Analytical Approach for the Optimal Proportioning of SMRF Buildings for Earthquake Resistance", at the EMI 2021 Virtual Conference, May 25-28, 2021. Click here to watch the video.
Seismic waves propagating into Los Angeles in an 1857-like simulated San Andreas fault earthquake.
Pre- and Post-Northridge Tall buildings, hypothetically located at Thousand Oaks, swaying (with one collapsing) under an 1857-like San Andreas fault earthquake.
Computationally Recreating an Experiment on Brace Buckling
Tall Building Response to Near-Source Ground Motion
Benchmark Problem for Structural Collapse
Toppling of Precariously Balanced Rocks in Earthquakes