Mohiuddeen Khan

Greetings! 👋 I am Mohiuddeen Khan, a recent graduate from the University at Buffalo, SUNY, USA where I earned my Master's degree in Computer Science & Engineering. Previously, I did my bachelor's degree in Computer Engineering from the Aligarh Muslim University, India. My interests lie in Software Engineering, Machine Learning, Blockchain, and Full-stack development.

I recently concluded a rewarding internship at TetraMem Inc. in the San Francisco Bay Area. In this role, I contributed to the development and deployment of MobileNetV1 and MTCNN machine learning models for in-memory computing AI inference chips. My responsibilities included addressing hardware-generated noise challenges through the incorporation of synthetic gaussian noise, ensuring optimal performance during chip testing.

Prior to TetraMem, I worked as a Software Engineer Intern at Samsung R&D in Noida, India where I implemented the TextRank algorithm to identify keywords and track user behavior, facilitating the activation of a secret mode in Samsung's SBrowser. Additionally, I played a key role in optimizing SBrowser's performance, thereby enhancing its overall quality. I also previously interned at the University of Tsukuba, Japan in the summer of 2021 where I worked under prof. Claus Aranha to estimate the behavior of different agents towards our player in the AI Werewolf game using a modified ensemble learning method. I also love to solve programming problems and do competitive coding on platforms like LeetCode and Spoj.

Currently, I am actively seeking full-time opportunities. If you have openings for Software Development/Full-Stack/Machine Learning roles, compelling projects, or if you simply wish to engage in discussions about the dynamic realm of computer science, I invite you to reach out to me.

Email  /  Linkedin  /  Github /  Google Scholar /  ResearchGate

Developed MobileNetV1 and MTCNN models to be run and deployed on in-memory computing AI inference chips.
Incorporated synthetic gaussian noise into the models to effectively handle hardware-generated noise during chip test.

Used TextRank algorithm to identify keywords from a webpage to track user behavior and activate secret mode when a page with related keywords was visited.
Developed scripts to store keywords in SQLite database and incorporated the algorithm in Sbrowser using Android.

Developed and applied a modified ensemble learning approach to predict the hostility and actions of other competitors in the AI Wolf game.
Made artificial agents to interact with other agents and play different roles in the AI Wolf game.

Published in the Mathematics Journal MDPI 2022.

Presented at the International Conference on Sustainable Advanced Computing (ICSAC 2021, Christ University, Bangalore) & published in Springer Lecture Notes in Electrial Engineering Conference Proceedings.

Published in proceedings (ACM) of the 4th International Conference on Machine Learning and Soft Computing (ICMLSC 2020) held in Hanoi, Vietnam

Submission in progress. Preprint available on arXiv.

Published in the IJSRSET Journal.

I have worked on various individual and group projects using various technologies . To see all of my projects please visit my Github profile. Some of my key projects are:-

Developed a tool to detect Cros-Site Scripting (XSS) attacks on websites using Ensemble learning approach.
Optimized and expanded the code to identify more types o malicious scripts, with the goal of making it open source for the developers to safeguard their websites.

Developed, tested, and tokenized a smart contract representing water as a digital asset utilizing the Ethereum ERC-721 tokenization standards, and used Remix to conduct transactions on an Ethereum test chain.
Utilized React.js to architect and develop a dynamic user interface for the WaterMarketPlace blockchain web-app.

Led development of Gesture-Controlled Mouse Interface using the Mediapipe library's convolutional neural network (CNN) for real-time hand tracking and gesture recognition.
Designed and implemented precise gesture detection algorithms, shaping a user-friendly interface and addressing accessibility.

Implemented the Alternating-Bit (ABT), Go-Back-N (GBN), and Selective-Repeat (SR) protocols for reliable data transfer among sending and receiving entities.
Wrote procedures for the transport layer for data transfer and compared the protocols in terms of different metrics.

Implemented a novel multi-model Deep Learning approach for Pneumonia detection using combination of Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and VGG-16 model.
Achieved a significant improvement in accuracy over individual models through the proposed hierarchical weighting strategy, showcasing its potential for binary classification tasks.

Implemented Programmable Logic Array into C++ code from scratch to properly implement various kinds of combinational logic circuits and devised an algorithm for solving the Quine–McCluskey method for the minimization of Boolean functions

Designed a software for predicting the safest route in Dhanbad city for women and children on Google Maps using Decision Trees and deployed the software as an emergency whatsapp number to show the route using Twilio whatsapp API. This project was done in the hackfest hackathon finale held at IIT Dhanbad.

Created a web based app for the Airports Authority of India to keep track of maintenance of equipments using predictive analysis and a portal to deploy engineers for a specific job. This project was done in the finale of Smart India Hackathon finale.

Website Template Credit to Jon Barron