Im a Software Engineer passionate about Machine Learning, Artificial Intelligence, Algorithms, and Backend Development. I love working on challenging problems especially anything related to optimization, autonomous systems, and data-driven intelligence.
- Advanced Machine Learning and AI Systems
- Algorithm Design, and Competitive, Programming (LeetCode enthusiast 🚀)
- Database Systems and Data Modeling
- Building end-to-end AI powered applications
- Programming Languages: Python, Java, C, SQL, JavaScript, Typscript
- AI/ML Frameworks: TensorFlow, Scikit Learn, OpenCV, Hugging Face Transformers
- Databases:
- SQL (MySQL, PostgreSQL)
- NoSQL (MongoDB, Firebase)
- Relational Algebra, 2NF/3NF Normalization, Schema Design
- Web/App Development: Streamlit, Flask, Node.js (basics)
- Tools: Git, GitHub, Docker, VSCode, Linux/WSL
- Deep Learning Architectures (CNNs, RNNs, Transformers)
- System Design for scalable AI applications
- Advanced SQL optimization - and indexing strategies
- Reinforcement Learning basics
-
Spatial-Semantic Mamba (SS-Mamba)
Developed a novel Mamba-based framework for whole-slide image (WSI) analysis that preserves 2D spatial structure and semantic relationships. Introduced 2D State Space Models (2D-SSM) with semantic patch ordering to address limitations of traditional MIL approaches.
Results: Achieved state-of-the-art performance on Camelyon16 (95.1% AUC), TCGA-BRCA (89.1% AUC), and BRACS (85.8% AUC).
📄 Graduate-level research project (unpublished manuscript / preprint) -
Facial Recognition Project
Built a real time facial recognition - system using OpenCV and Deep Learning. Implemented face detection and recognition pipelines with efficient matching algorithms. -
Linear Regression Project
Designed and trained a linear regression model from scratch to-predict trends based on real-world data. Focused on gradient descent optimization and loss minimization. -
Neural Network Project
Built a custom neural network model-without using high-level frameworks. Implemented forward pass backpropagation, and trained on classification datasets, Including Cosine Similarity checking and Euclidean Distance. -
SQL Database Manager
CLI tool built with Java and JDBC for library database management complete CRUD functionality, SQL queries and database normalization up to 3NF. -
AutoSeedBot
Autonomous ecofriendly robot designed for reforestation. Implemented sunlight detection for navigation and automated seed distribution using custom-coded 3D-printed hardware.