Hi, my name is

Charitarth.

I craft intelligent systems

An ambitious undergraduate student with a passion for Machine Learning and its applications. I strive to delve into the latest advancements in the field, eager to contribute to innovative solutions and uncover the power of data-driven intelligence.

Resume

About Me

I am a passionate and ambitious undergraduate student with a solid foundation in Computer Science and Statistics. My keen interest in Machine Learning drives me to explore the intersection of technology and creativity, where data-driven intelligence can solve real-world problems. Currently, I am immersing myself in innovative solutions within this field, eager to contribute to its advancements and make a meaningful impact. My journey is fueled by a desire to leverage technology for transformative and practical applications, and I am excited about the opportunities ahead to grow and collaborate with like-minded professionals. Here are a few technologies I've been working with recently:
  • PyTorch/ PyTorch Lightning
  • Albumentations
  • Transformers
  • Scikit-Learn
  • Pandas/Polars
  • FastAPI

Experience

Freelance Machine Learning Engineer - IOPS
Aug 2023 - present
I am currently working as a freelance Machine Learning Engineer at IOPS on a project that uses decentralized federated learning to detect network attacks.
REU Researcher - University of Connecticut
Jun 2023 - Dec 2023
Worked at the Connecticut Transportation Institute building autonomous systems for the road.

Education

2021 - 2025
Bachelors of Science in Mathematics-Statistics, Minor in Computer Science
University of Connecticut

Relevent Coursework:

  • Deep Learning, Data Structures & Algorithms, Systems Programming, Cybersecurity, Machine Learning, Data Science

Extracurricular Activities:

High School
Trumbull High School

Projects and Research

Energy Justice Mapping Tool
Python Pandas GeoPandas
Energy Justice Mapping Tool

Project Overview: Empowering communities by identifying energy deserts and promoting equitable access to energy infrastructure.

Key Achievements:

  • Collaborated with a multidisciplinary team to craft a winning proposal for the Clean Energy & Sustainability Innovation Program 2024, securing $2,500 in funding and recognition from industry leaders.
  • Presented to White House officials and directors of Eversource Energy at the Clean Energy Summit 2024, showcasing the tool’s potential impact on energy policy and infrastructure.
  • Paper submitted to IEEE PESGM 2025, further advancing the field of energy justice and sustainability.
Link
LightningSparseInst
Python PyTorch PyTorch Lightning Weights and Biases FiftyOne Albumentations
LightningSparseInst
Replicating the results of Sparse Instance Activation for Real-Time Instance Segmentation by Cheng et al. (2022), published at CVPR 2022. Implementing SparseInst on ResNet-50, and investigating performance with other efficient backbones.
Made data loading on COCO dataset 3x faster than default using caching and fast JSON parsing. Implemented custom transforms in Albumentations and a Custom Dataset Exporter in FiftyOne.
Encoding high-dimensional court data for pre-trial motion outcome prediction
PyTorch Lightning Pandas
Encoding high-dimensional court data for pre-trial motion outcome prediction
Working under Dr. Derek Aguiar to predict outcomes of motions. Implemented a custom multimodal predictor that achieves state of the art results on data from the Connecticut Judiciary.
Certfied Adversarial Robustness With Parameter Efficitent Fine Tuning
Certfied Adversarial Robustness With Parameter Efficitent Fine Tuning

Working under Dr. Derek Aguiar to improve the robustness of vision models against adversarial attacks.

Image is from “Towards Deep Learning Models Resistant to Adversarial Attacks” by Madry et al. (2017)

Achievements

HackHarvard Best Efficiency Booster Hack
Awarded for SnipStudy, a platform for gaining insights from long form content.
HackUMass Best Use of Twilio Hack
Awarded for Who's There, a smart home security notification system

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!
Mail me