Mann Sapariya
@mannsapariya
Mann Sapariya
@mannsapariya
Mumbai, India
Mann Sapariya
๐ About Me
Hi! I'm Mann Sapariya, a passionate Machine Learning & Full-Stack Developer with a strong background in AI, Deep Learning, and Web Development. I love building intelligent solutions and working on real-world applications of technology. Currently pursuing B.Tech in Information Technology at Dwarkadas J. Sanghvi College of Engineering (2022-2026) with a CGPA of 8.22.
๐ Mumbai, India๐ง [email protected]๐ GitHub
๐ Skills
๐ง Machine Learning & Deep Learning
Algorithms: Linear & Logistic Regression, Decision Trees, Random Forest, KNN, K-Means
Frameworks: PyTorch, TensorFlow
Computer Vision: OpenCV, EfficientNet, ResNet, MobileNet, Vision Transformers, EfficientNet with RNN
NLP: Hugging Face (BERT, RoBERTa, mBERT, XLM-R), NLTK
Statistics & Data Analysis: Probability, Hypothesis Testing, Bayesian & Frequentist Inference
๐ป Programming & Web Development
Languages: Python, SQL, Java, C, C++, JavaScript, HTML, CSS
Backend: Flask
Data Visualization: Tableau
Soft Skills: Teamwork, Communication
๐ผ Professional Experience
๐น Datamatics TruCap+ (Project Trainee) | Feb 2025 - Mar 2025
Working on LLMs like DONUT, LayoutLMv3, LLaMA, Mistral 7B for invoice & receipt processing.
Developing deep learning models for structured data extraction from financial documents.
๐น AIESEC in Mumbai (Junior Manager - Incoming Corporate Sector) | Jan 2025 - Aug 2025
Led contract negotiations with corporate clients, driving strategic partnerships.
Managed CRM data, ensuring accurate client records for outreach via Email & LinkedIn.
๐ Achievements & Hackathons
๐ Winner - AWS i-Hack 25 (IIT Bombay)๐ 1st Prize for AI-powered financial fraud detection models (Spam Messages, Fraud Calls, Deepfake VKYCs).
๐ฅ Winner - Tech-a-thon๐ 2nd Prize in a 24-hour hackathon for developing an urban planning full-stack platform with GenAI-based analysis, predictions, and recommendations.
๐ Finalist - PICT TechFiesta 2025๐ Built a unified healthcare dashboard in a 24-hour hackathon.
๐ Winner - PLANNERED 4.0 Hackathon๐ 1st Prize for an agro-tourism business model & tech solution.
๐ Notable Projects
๐ DeepFake Detection
๐ Achieved 99.81% accuracy using a ViT 224 model trained on 34,000 images.
๐ฆด Bone Fracture Detection
๐ฉบ Deployed ResNet-50 model with 97% train accuracy & 80% test accuracy on a dataset of 4,000 images.
๐ก CAPTCHA Detection
๐ Built a PyTorch model with EfficientNet + RNN, trained on 1,700 samples for high-precision decoding.
โ๏ธ Flight Price Predictor
๐ Achieved 91% precision using Linear Regression & data analysis.
๐ Stock Market Analysis
๐ Analyzed Kaggle stock market data & visualized trends.
๐ Sentiment Analysis Research
๐ Conducting a comparative study on multilingual sentiment analysis (English & Hindi) using BERT, XLM-R, T5, XLNet, GPT-3, LSTM, SVM, LazyClassifier.
๐ Certifications
โ AWS Academy - Machine Learning Foundationsโ AIESEC in Mumbai - Letter of Recommendationโ Course Completion Certificates
๐ค Let's Connect!
๐ GitHub๐ง [email protected]๐ Mumbai, India
๐ฏ Open to collaborations, research projects, and internship opportunities! ๐