SAHIL KOMMALAPATI

Logo

View My GitHub Profile

Hello!

I’m a Ph.D. student at the University of Texas at Austin, where I’m advised by Prof. Robert Moser. I work at the intersection of Fluid Turbulence and Machine Learning. I was previously a student researcher at Google where I worked on physics-based deep learning approaches for improving weather predictions. Before that, I was a Machine Learning predoc at Argonne National Laboratory where I worked on developing uncertainty quantification methods for deep nerual networks with a focus on physics predictions.

If you are looking to work under my guidance for your undergraduate or graduate projects, connect with me over LinkedIn!

News

[Jun 2023] I joined Google as a Student Researcher!

[Aug 2022] I joined UT Austin to begin my PhD in ME 🥳🥳

[Nov 2021] I joined Argonne National Laboratory as a Predoctoral Researcher in Machine Learning.

[Jun 2021] I joined Telepath AI as a Data Science Research Fellow.

[Jun 2021] I graduated from the University of Washington with an MS in Mechanical Engineering.

[Jun 2021] I defended my MS thesis on “Machine Learning for Coherent Structure Identification and Super Resolution of Turbulent flows”.

[Dec 2020] I received the Herbold Data Science Fellowhip.

[Nov 2020] I presented my research on turbulent coherent structure identification at APS-DFD 2020.

[Dec 2019] I joined Williams Turbulence Laboratory as a graduate student researcher.

Videos featuring my work:

  1. Talk at 2020 APS-DFD, introducing my thesis research on Bayesian optimization driven coherent structure identification. Video Link.

  2. Research presented at the 2020 NCMDAO conference hosted by Vikram Sarabhai Space Center. Video link.

  3. My Lectures for the Workshop series on Machine Learning in association with Mechanical Engineering Graduate Students association (MEGA, UW). Playlist Link.

Selected projects from my previous experiences:

  1. Super-resolution of turbulent particle Image velocimetry datasets using Machine learning. GitHub Link.

  2. Using Bayesian Analysis to identify coherent structures in Turbulent boundary layer data. Here is a presentation.

  3. Using Model Predictive Control to stabilize the wake of a Rotating Cylinder. We are using Dynamic Mode Decomposition (with control) to linearize the flow model.Github Link

  4. My review of the Koopman Operator theory for systems in the continuous spectrum. PDF

  5. My Bachelor’s thesis on the stability and Control of an Intercontinental Ballistic Missile. PDF

Published original research:

Check out my Google Scholar or reach out on ResearchGate.

Other works online:

I truly believe that elegant and clear teaching is a skill that is fundamental to the overall learning process and I thoroughly enjoy it. It strengthens my authority and confidence over the specific material and it ultimately nurtures my abilities towards better scientific communication. So, in that spirit, I have picked up the habit of writing about things that I find exciting and impactful. Visit my Medium page on the left (or top, if you are on your phone) to check out some of my articles.

website theme by orderedlist.