Hiral Nagda

Hiral Nagda

San Francisco Bay Area
2K followers 500+ connections

Experience

  • Meta Graphic

    Meta

    Menlo Park, California, United States

  • -

  • -

  • -

Education

  • Rutgers University Graphic

    Rutgers University

    3.806/4

    -

    Coursework: Introduction to Artificial Intelligence, Introduction to Data Structures and Algorithms, Topics In The Foundations Of Computer Science, Numerical Analysis, Topics of Computers in Bio-medicine, Massive Data Mining, Pattern Recognition, Massive Data Storage and Retrieval, Computer Vision, Design and Analysis of Data Structures and Algorithms

  • -

Licenses & Certifications

Volunteer Experience

Projects

  • Movie Recommendation System.

    Implemented movie recommendation system using K-Nearest-Neighbor, Matrix-Factorization and Slope-one algorithm on Movielens dataset.

  • Colorizer

    • Developed a model which converts grayscale images to color images. Used 3 × 3-pixel window of gray values, and mapped this set of nine gray values to a single (r, g, b) color vector.
    • Designed a 5-layer Artificial Neural Network architecture using Keras.

  • Johnson's Algorithm Snippet

    Visualized working of Johnson's Algorithm using networkx and matplotlib

  • Minesweeper

    The goal of this project is to design an AI that plays MineSweeper game - that is, a program capable of sequentially deciding what cells to check, and using the resulting information to direct future actions.

  • Mazerunner

    •Deployed an Intelligent System to solve randomly generated maze using BFS, DFS, and A* algorithm.
    •Implemented Genetic Algorithm to build hard-to-solve mazes.

  • Haar Cascade

    Created Haar Cascade for face, eyes and object detection (Watch) using OpenCV, implemented on Google Cloud Instance.

  • Object Detection System

    •Classification and localization of 80 classes of objects using deep learning algorithm YOLOv2.
    •Filtered threshold on class scores. Used second filter Non-Max Suppression.
    •Used deep CNN of the 19x19x5x85 dimensional encoding.

  • Insurance Policy Prediction:

    -

    •Used Recursive Feature Elimination method using logistic regression model to find best
    features.
    •Formed K-means clusters using different attributes for analysis.
    • Deployed Decision Tree (89% accuracy) and Random Forest (89.125% accuracy) models for training of the model.
    •Performed K-fold cross validation and hyperparameter tuning to choose the best model.
    • Generated User Interface for the recommendation system

  • E-Library Management Android Application

    -

    •Developed an application using live database of computer department library.
    •The application provided features for students to check for availability of a book and consequently request for the same.
    •The student is notified when they cross deadline of returning book.
    •Technologies used: REST API, Volley Library, Android
    Studio, PHP, MySQL, Java, XML.

  • 2D Racing Game

    -

    Implemented 2D Racing game for Android phones on Unity using C#. Also generated user interface and
    sprites for the game.

Test Scores

  • GRE

    Score: 320

    Verbal-150
    Quants-170

Recommendations received

View Hiral’s full profile

  • See who you know in common
  • Get introduced
  • Contact Hiral directly
Join to view full profile

Other similar profiles

Explore top content on LinkedIn

Find curated posts and insights for relevant topics all in one place.

View top content