8000 Dhawal-Modi (Dhawal Modi) · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content
View Dhawal-Modi's full-sized avatar
🎯
Focusing
🎯
Focusing

Highlights

  • Pro

Block or report Dhawal-Modi

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Dhawal-Modi/README.md

Typing SVG

I'm currently pursuing a Master of Science in Electrical Engineering & Computer Science (MS EECS) at the University of California, Merced.

Before starting my MS, I completed my Bachelor of Engineering in Electronics & Communications Engineering at RGPV-UIT in India. During my undergraduate studies, my coursework included , Digital Circuits, Design & Analysis of Algorithms, Internet of Things, Computer Architecture, Computer Networks, Operating Systems, Software Engineering, Statistics, Microwave Engineering.

Work Experience:

In terms of work experience, I had the opportunity to work as a Backend Software Engineer at Tata Consultancy Services Limited. I contributed to integrating various payment settling systems like Direct Debit, Credit Transfers and E-Mandates, streamlining payment processes, reducing transaction settlement times. Moreover, I orchestrated the deployment of mock webservices using Spring Boot applications using Scrum, Agile Methodology, Java, and Maven.

I also am currently working as a Graduate Student Researcher in UC Merced's MoCA Lab . My work in the lab includes developing and optimizing Deep Learning vision algrorithms for use on the Nvidia Orin AGX embedded platform.

Projects:

A project I'm proud of is "Efficient Person Following Robot algorithms", where I am developing an architecture for person following and tracking using Pose estimation and Person detection. I am currently optimizing current SOTA methods to work efficiently on embedded platforms like Nvidia Orin AGX.

Another recent project that i completed was LLM Prompt prediction. The goal of this project was to predict a prompt, ⟨Rewrite Prompt⟩, that was used to generate ⟨Rewritten Text⟩ by providing only ⟨Original Text⟩-⟨Rewritten Text⟩ pairs to the LLM. At inference, it predicts an embedding using the ⟨Original Text⟩ and the ⟨Rewritten Text⟩. The finetuned models ranked 986 out of 2186. The best leaderboard score was our Phi-2 submission which scored 0.6188. The Mixtral 8x7B model finetuned with 1000 samples scored 0.6056 and Gemma 7B finetuned with 5000 samples scored 0.5981.

Technical Skills

  • Programming Languages & Databases: Python, Java, HTML, CSS, JavaScript, C, C++, SQL, Postgre SQL, Oracle SQL, IBM Db2
  • Tools: Git, Jenkins, Docker, Colab, Maven, Jupyter
  • Frameworks: Express.js, ROS, Spring, Spring Boot, FastAPI
  • ML Frameworks: PyTorch, Seaborn, Scikit-learn, NumPy, Pandas, Keras, TensorFlow

Hackathon Experience:

  • Won 2nd prize (worth $5000) in Sentient's Werewolf AGI-thon for developing a Llama 2 70B agent capable of playing the social deduction game of Werewolf.
  • Won 1st prize nationwide in Smart India Hackathon 2018 (Hardware Editon) for developing a Computer Vision based real-time road traffic mangement system.
  • Was selected in top 5 teams natiowide in TCS Finorama Fintech challenge 2018.

Extracurricular Activities:

  • Served as a Teaching Assistant for CSE 015: Discrete Mathematics at UC Merced, where i conducted and managed labs for 90 undergradute students.

Please feel free to get in touch with me, I'm excited to connect and collaborate in Software developement, machine learning, and beyond.

My GitHub Stats



Shreya's Activity Graph

📫 How to reach me:

dmodi2@ucmerced.edu LinkedIn

Show ❤️ by starring repositories you find good!

Pinned Loading

  1. llm-prompt-recovery llm-prompt-recovery Public

    Forked from kevin-chau/llm-prompt-recovery

    Repository for LLM Prompt Recovery Project.

    Jupyter Notebook

  2. DSC_2024_Space_Jam DSC_2024_Space_Jam Public

    Jupyter Notebook 1

  3. Kaggle_Playground_S4E01 Kaggle_Playground_S4E01 Public

    Jupyter Notebook

  4. Kaggle_Playground_S4E02 Kaggle_Playground_S4E02 Public

    Jupyter Notebook

  5. Mini-Weather-Reporting-API Mini-Weather-Reporting-API Public

    This Node.js script pulls JSON weather report using the OpenWeatherMap API and prepares it into a nice HTML Report.

    JavaScript

  6. Dynamic-Traffic-Management-using-OpenCV Dynamic-Traffic-Management-using-OpenCV Public

    Code for 'Dynamic Traffic management system'.

    C++

0