π I'm an undergraduate studying Computer Engineering at the University of Illinois Urbana-Champaign (UIUC).
π¨βπ» I'm experienced in and deeply passionate about machine learning, space/astronomy, and robotics, and I'm eager to continue diving into the applications of intelligent systems, especially for outer space environments!
π I'm currently working as a computational astrophysics research intern for the Champaign Supernova Team and the NCSA Gravity Group at Illinois' National Center for Supercomputing Applications (NCSA), as well as leading UIUC's lunar robotics team, Illinois Robotics in Space, to compete in the NASA Lunabotics Challenge.
π§ For more information on my skills, projects, and experience, see my work below, check out my website at https://arjunchainani.github.io/, or contact me through email at arjun.k.chainani@gmail.com!
In addition to my pinned personal projects, here are some other open-source, large-scale projects I've made contributions to that GitHub doesn't let me pin for various reasons (e.g. is a fork from a previous open-source project or commits were made from another account):
- LSSTDESC/RESSPECT - The Recommendation System for Spectroscopic Follow-Up, a framework for filtering through several terabytes of nightly photometry data from the Legacy Survey of Space and Time, and identifying transients that merit detailed but expensive spectroscopic follow-up. Its active learning nature allows the system to learn the key underlying features behind supernova light-curves, and adapt from initial simulated training data to noisy, real-world telescope data that likely changes throughout the 10-year survey.
- LSSTDESC/oracle-resspect-classifier - Implementation of the state-of-the-art ORACLE (the Online Ranked Astrophysical Class Estimator, Shah et. al) hierarchical RNN classifier within RESSPECT to allow the aforementioned framework to identify a wider range of transients more accurately and earlier in the event life cycle than ever before. This work also augments the capabilities of RESSPECT to perform anomaly detection tasks and identify previously rare or unseen objects with a large potential for novel scientific discovery!
- IllinoisRoboticsInSpace/IRIS - Codebase for the autonomous lunar rover we design on the Illinois Robotics in Space project team. Written with C++, Python, and Bash as part of the ROS2 framework. Contains modules and branches for teleop control systems as well as higher-level autonomy algorithms.