About
Experience
- Customized a tree-based optimization algorithm that separates drugs into distinct groups to maximize the concurrent execution of drug dispenser robots; improved an optimization algorithm that reduced the cycle-time of the automation process by 25%.
- Devised a customized recursion algorithm to extend the functionality of the current environment to a multi-robot and multi-system environment
- Built a modular video analytics app using Flask, OpenCV that tracks the location of each human in a multi-camera environment; utilized perspective transformation, object detection, and object tracking to find the location of a human.
- Increased the speed of the video analytics app by 20% by using a customized YOLOv3 algorithm to perform object detection and DeepSORT algorithm for multi-object tracking.
- Tools: Python, Flask, OpenCV, Keras, Tensorflow, PyTorch
- Programmed a cloud-based web app using Flask and JavaScript aimed at storing, visualizing, and tracking temperature and humidity of each drug container to estimate the degradation of the drug.
- Created an alert system to send notifications and emails when the parameters exceed the threshold.
- Tools: Python, Flask, JavaScript
- Designed a library management system for a local library, where we undertook activities like requirement elicitation, preparing Data Flow and Entity-Relationship diagrams.
- Delivered a solution for a POC involving Automatic Financial Document Classifier using Natural Language Processing and Support Vector Machines with 96% accuracy on the company’s data.
- Tools: Python, Scikit-learn, NLTK
Projects
A music streaming web app based on Django
- Tools: Django, HTML, CSS, Bootstrap, SQLite, AWS S3, Heroku
- Register/login to the web app(with OAuth-based Google Sign-In).
- Search and filter songs based on language and singer.
- Create multiple playlists and add/remove songs to/from playlist.
- Scroll through recently played/viewed songs.
An attention-based classification model that aims at generating an answer for a given input image.
A Seq2Seq model that generates a short summary of the given input video.
An image generator based on the concept of adversarial networks (GANs)
Skills
Languages and Databases
Libraries
Frameworks
Other
Education
School of Engineering and Applied Science, Ahmedabad University
Ahmedabad, India
Degree: Bachelor of Technology in Information and Communication Technology
CGPA: 3.13/4.33
- Data Structures and Algorithms
- Database Management Systems
- Operating Systems
- Machine Learning
- Computer Vision
Relevant Courseworks: