Chicago, Illinois, United States
Data Scientist specializing in AI/ML and High-Performance Computing (HPC) at the University of Chicago's Research Computing Center (RCC). I collaborate with researchers to develop and implement cutting-edge solutions, leveraging LLMs, deep learning, and HPC resources. My work spans diverse domains, including materials science, climate modeling, and urban studies. See my projects on GitHub!
I worked on emotion recognition in text and speech. I implemented and analyzed BERT-based techniques and contrastive learning pipelines on IEMOCAP and MELD datasets
I worked with Prof. Jeff Cohn in the field of psychology and machine learning. Specifically, I worked on diarization and on the alignment of video, audio, and text data from Deep Brain Stimulation and other mental health datasets. I also worked on turn-taking in text using BERT for human-computer conversations from the DAIC dataset.
I worked as a Graduate Research Assistant, advised by Dr. David Wettergreen. My work entailed extensive use of machine learning and especially deep learning to study and analyse orbital data, specifically hyperspectral image cubes. I also worked on the software development for the Zoe robot, Life in the Atacama project which involves experience with ROS and the use of the Mars Microbeam Raman Spectrometer.
I was a TA for 10701 for the Spring 2020 semester taught by Prof. Tom Mitchell and Prof. Leila Wehbe. It is an introduction to machine learning course for PhD students.
I worked as a Teaching Assistant for 16681 MRSD Course Project for Spring 2019, instructed by Prof. John Dolan. The course provided a platform for students to work on systems engineering projects and learn how to build and integrate various robotics sub-systems successfully.
◦ Worked under Prof. Majid Zamani at the Hybrid Control Systems (HCS) Lab ◦ Implemented real-time object detection and User Datagram Protocol (UDP) based communication algorithm for communication between a processing computer server and several robot clients (Khephera-IV robots) ◦ Designed and implemented state feedback controller to control the motion of the robot from one defined state to another and experimented with SCOTS (toolbox developed at HCS for synthesizing symbolic controllers) and implemented equations defined in linear temporal logic on Khepera-IV robots