My first coding project, back in the ninth grade was to build a chatbot. It was a rather tall ladder of if-else statements nested within if-else statements. Nevertheless, it was enthralling to be able to chat with a machine.
Six years later, I recreated my chatbot once again, only with a lot more knowledge! This time, it was an intelligent agent focused on thinking and acting like a human as well as thinking and acting rationally.
I'm a Masters of Technology student majoring in Computer
Science at MSRIT, India. I have spent the last six years developing my skills in computer science to be helpful to our world. My skill set is a diverse palate of computer science technologies. I've built several
end-to-end web and android applications, a smart vending machine prototype, and analysed, optimised and designed several algorithms to solve problems for the people around me. After all, I discovered my supreme interest in
artificial intelligence, machine learning, deep learning and natural language processing.
I've worked on using machine learning and NLP on legal research to ease the arduous task of researching and reading lengthy judgments using a conditional random fields model and built classification models and neural networks
to classify judgements based on their final decision. I worked on models for speech pronunciation correction in Arabic. Currently, my research work is on finding better solutions for computational problems using machine intelligence.
In my spare time, I read and learn about psychology, language, history and culture. I study the art of Japanese flower arrangement at the Sogetsu Schook of Ikebana. I love having deep and meaningful conversations about
life.
Here's my linkedIn, resume and github.
What I can do.
I can help.
If you have a project that my skill-set would be useful to, think you need my help with something or just fancy saying hello, then get in touch.
A Decentralized AI Marketplace (2021-2022)
Technologies: Machine Learning, Ethereum, Secure MPC , Python , React , Next.js, Node.js, Solidity, Ganache
Artificial Intelligence and blockchain are two disruptive technologies proving to be a powerful combination, improving virtually every industry. In the final year of my MTech in 2022, my research focused on the amalgamation of these technologies. We created a Blockchain-based marketplace for AI services. The ecosystem has three components to understand it: Publishing AI Services, Contribution to AI Services, and Fetching Predictions from AI Services. The first component allows corporations, organizations or individuals to publish an AI service that solves any problem statement. An AI service publisher can opt to publish their service in a ‘private’ or ‘public’ mode. In ‘public’ mode, the dataset and model will be put up for sale at the price (in ether) as listed by the publisher. Every purchase of the dataset or model would increase the reputation of the dataset or model, and in turn, increase the sale value of the dataset or model. For instance, a dataset, originally listed for 40000 wei has been purchased 20 times. With every purchase, there is an increase in the reputation ranking and there is a price increase of 100 wei. Therefore, following 20 downloads, the price of the dataset will be 42000 wei. In ‘private’ mode, the dataset and model will not be put up for sale but rather securely deployed to serve purchased predictions for requested input data. Secure multi-party computation is the foundation of serving predictions in ‘private’ mode. Notably, when the AI Service Publisher solves for a particular problem statement, they create an open thread for that particular problem statement.
The second component allows machine learning practitioners to contribute to these problem statements that AI service publishers have published a service for. Contributors can contribute datasets or models for this problem statement. As described before, every purchase of the dataset or model would increase the reputation of the dataset or model, and in turn, increase the sale value of the dataset or model. Consequently, users will be able to identify better-quality datasets and models. Further, to validate a dataset, 10% percent of the data points are revealed to users. Also, to validate the model, a demo API of 5 calls are set up for that model. Thus, the ecosystem allows corporations, organizations and individuals with problems that could be solved with AI to solicit solutions from AI Service Publishers and Contributors. The third component allows users to purchase datasets, models or predictions from the contributors or AI service publishers.
Artificial Intelligence for Legal Research (2019)
Technologies: Machine Learning, NLP, Deep Learning, Python
The project was built to ease the arduous task of researching and reading lengthy judgments. To help find the required information at a glance, the text in a judgment document was categorized with the help of Conditional Random Fields (CRF) to fall into one of the six predetermined categories. The training data consisted of documents from the rent control domain, and the sentences in these documents were annotated with the category they belonged to. The model was trained with four sets of features to help categorize sentences in other judgments. Further, these documents were classified based on the final decision taken by the judge with the help of a neural network classifier.
An Angular Annotation Tool (2018)
Technologies: Angular6, Python, Cloud Firestore
Developed an annotation tool that provides an interface to a sentence-segmented corpus making it easy to annotate sentences with tags.
Technical fest: DexteriX’18 Website (2018)
Technologies: HTML5 , PHP, Javascript, jQuery, Mysql
Developed a website that handled 2500+ registrations and provided complete information on all the inter-collegiate events held at the fest.
Computer Graphics App. On Routing Information Protocol (2018)
Technologies: C, OpenGL
Developed using OpenGL to demonstrate the working of Routing Information Protocol.
Arabiyah Classes Android Application (2017)
Technologies: Java, Android
Developed an android application for an Egyptian institute that allows users to enroll and purchase courses as well as provide information on the institute and its courses.
Weather Forecast And News Android Application (2017)
Technologies: Java, Android, REST API
Developed an android application that serves the user with updated weather forecast as well as updated and accurate news from global news networks.
Department Management System Web Application (2017)
Technologies: HTML5 , PHP, Javascript, jQuery, Mysql
Developed a web application that processes all student and faculty details, academic-related reports, course details, curriculum, batch details and other resource-related details.