Hello, my name is

Darshan Rao

Bridging Research and Innovation

I am a Computer Science graduate student at USC with a keen interest in machine learning and natural language processing. I am currently engaged in research at INK Labs, USC , where my focus lies in developing fact verification systems. Additionally, I am involved in creating benchmark datasets that highlight domains where Large Language Models tend to generate inaccurate information.

About Me

I'm Darshan, a grad student at University of Southern California majoring in Computer Science. My journey began with a degree in Computer Engineering from the University of Mumbai, where I developed a passion for building mobile apps with Flutter. My interest in machine learning, especially neural networks, took off during my research internship at TIFR. I've always been more drawn to solving real-world problems than sticking to just academic theories.

Currently, at the INK Lab at USC, I'm investigating how large language models function and the sensitivity at which they hallucinate. I'm excited about creating benchmarks to better evaluate these models.

With a long-standing passion for software development and machine learning, I'm eager to make my mark in this field and contribute to its future.

Here are the list of technical skills I have:

  • Languages
    • Python
    • C++
    • SQL
    • Bash
    • Dart
  • Libraries & Frameworks
    • TensorFlow
    • PyTorch
    • LangChain
    • Scikit-Learn
    • Pandas
    • Flutter
    • React
  • Cloud Services & DevOps
    • Amazon Web Services
    • FireBase
Headshot

Where I’ve Worked

Research Assistant @ INK Lab, USC

Jun 2024 - Present

  • Develop comprehensive framework that uses large language models (LLMs) for automated fact-checking to evaluate factual accuracy
  • Establish consistent benchmarks and criteria to ensure fair and reliable comparison of LLMs’ factuality from multiple perspectives

Some Things I’ve Built

Other Noteworthy Projects

view the archive
  • Folder
    GitHubExternal Link

    Sequence Alignment Using Dyanamic Programming & Divide & Conquer

    This project implements and compares dynamic programming solutions for the Sequence Alignment problem, including a memory-efficient version, providing alignment costs, optimized sequences, performance metrics, and insightful analysis.

    • Python
  • Folder
    GitHubExternal Link

    Named Entity Recognition with BiLSTM

    This project implements a Named Entity Recognition (NER) system using a Bidirectional Long Short-Term Memory (BiLSTM) neural network. The system is designed to read in a dataset, train a BiLSTM model on this data, and then use the trained model to predict entity tags for unseen text.

    • Python
    • PyTorch
    • NumPy
    • Scikit-Learn
  • Folder
    GitHubExternal Link

    HMM-Based POS Tagger

    This project implements a Part-of-Speech (POS) tagger using a Hidden Markov Model (HMM). It includes code for training the HMM on a dataset, creating a vocabulary, and using two decoding algorithms (Greedy and Viterbi) for POS tagging.

    • Python
    • JSON
  • Folder
    GitHubExternal Link

    Sentiment Analysis Using Word2Vec

    This project is an extension of a previous assignment on sentiment analysis, now incorporating advanced neural models and word embeddings. The focus is on employing deep learning techniques to enhance the classification accuracy of sentiment analysis on Amazon kitchen product reviews.

    • Python
    • Pandas
    • Scikit-Learn
    • PyTorch
    • Gensim
  • Folder
    GitHub

    Little Go Game AI Agent

    This Go game AI agent is implemented in Python using the Minimax algorithm to play the game of Go. The agent considers various game states and employs strategic decision-making. This is a homework assigment for CSCI561 Foundation of Artificial Intelligenc

    • Python
  • Folder
    GitHub

    Genetic Algorithm for Travelling Salesman Problem

    This project is a Genetic Algorithm solution for the Travelling Salesman Problem (TSP). This is a homework assigment for CSCI561 Foundation of Artificial Intelligence

    • Python

What’s Next?

Get In Touch

As I actively seek full-time and co-op opportunities starting in Spring 2025, I welcome valuable connections and conversations. Whether you have insights to share, opportunities to discuss, potential collaborations, or simply want to say hello, my inbox is always open. I appreciate your time and look forward to engaging with you!