Angad Singh | AI/ML & Web3 Explorer

Hi, I'm Angad Singh.

A

Self-driven, quick starter, passionate programmer with a curious mind who enjoys solving complex and challenging real-world problems.

About

I am a final-year B.Tech student at Vellore Institute of Technology (VIT), with a passion for problem-solving and coding. I always strive to bring 100% effort to everything I do. I’m an AI/ML and Web3 enthusiast, passionate about solving complex problems and building innovative solutions. As a versatile developer specializing in Backend, AI & ML, and Web3 Development, I am driven by a passion for continuous learning. I enjoy exploring new technologies and creating user-focused, impactful applications.

Looking for an opportunity to work in a challenging position combining my skills in Software Engineering & related fields, which provides professional development, interesting experiences and personal growth.

Experience

SmartInternz

AI/ML Intern

Innovated a healthcare data analytics platform by integrating data sources, optimizing backend workflows, and implementing efficient ML algorithms.
Improved data processing efficiency by 30% and improved disease detection accuracy to 85%+ by streamlining ETL processes and optimizing SQL queries.
Collaborated with front-end developers to design a user-friendly interface using JavaScript, integrating custom DL models for disease detection, resulting in a 20% increase in user engagement and enhanced platform versatility, employing Agile methodology.
Tools: Python, Flask, MySQL

August 2023 - November 2023 | Remote, India

Projects

                
AlgoMentor
                  AI-Powered Socratic Educational Platform.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, React, Node.js, Express.js, FastAPI, MongoDB
Developed a Socratic AI tutor with RAG, multimodal AI, and real-time DSA visualizations, improving accuracy by 80% and debugging by 40%.
Automated 90% of queries, boosting engagement by 60% for students.

Boolean Networks
                  Visualize Boolean Functions using Customizable Neural Networks.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, React, Flask, numpy, pandas, tensorflow, matplotlib
Neural network training for Boolean functions with multiple activations and loss functions.
Lightweight AI deployment with PCA-based visualization on low-resource environments.

Sorting Hat
                  Sorting Hat from the Harry Potter Universe, made using Flask.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, Flask, numpy, pandas
Created an interactive ML-based personality classification system for user engagement.
Built a Gaussian Classifier to sort individuals into Harry Potter houses with 81.34% accuracy using Big Five Personality Scores.

ExamSecure
                  Blockchain-Powered Facial Recognition Application.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, Solidity, React, Node.js, web3.js, Python, Flask, OpenCV, FaceDB
Developed a secure online exam platform with real-time proctoring, achieving ~99% accuracy.
Enhanced exam integrity by 40%, providing a reliable solution for secure remote assessments.

MindSyncPlus
                  AI-Powered Healthcare Platform, designed using Flask.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, Flask, MySQL, Python, Flask, Tensorflow, Keras
Developed a healthcare platform for detecting Alzheimer's stages using ML models, achieving 85%+ accuracy.
Enhanced data efficiency by 30% and user engagement by 20% with a streamlined interface for Users.

Merkle Tree Verifier
                  A DApp for verifying transaction inclusion in a Merkle tree on Ethereum.
                
AccomplishmentsTools: HTML5, CSS3, JavaScript, Solidity, React, Node.js, web3.js, Flask
Developed a Solidity-based Merkle Proof Verifier smart contract, reducing transaction verification time by 30% with minimal gas costs.
Improved scalability by 20%, enabling secure and efficient handling of 10,000+ transactions on Ethereum.

Genome Classifier
                  Genome Classifier, built using Python. 
                
AccomplishmentsTools: Python
Built a genome classification model with Multinomial Naive Bayes and NLP, classifying species like chimpanzees, humans, and dogs.
Achieved high accuracy rates: 99.8% (chimpanzees), 98.6% (humans), and 92% (dogs), showcasing its value for genetic research.