I am building a startup that develops cutting-edge software solutions leveraging machine learning, IoT, and advanced pattern recognition. Our goal is to deliver scalable and impactful tools that address real-world challenges across industry, research, and infrastructure sectors.

• Uninet: One Framework, All Neural Tasks Uninet is an open-source framework designed to unify a diverse range of machine learning tasks—including classification, regression, clustering, and physics-informed modeling—within a modular, extensible architecture. It provides reusable components and templates to accelerate model development and deployment.

• WeatherWise
  Automated ML pipeline for rainfall prediction using FastAPI, DVC, and GitHub Actions for CI/CD, data versioning, and efficient model inference.

• Physics-Informed Neural Network
  Framework with FastAPI for solving PDEs (SHM, Heat, Wave, Burgers' equations) using Physics-Informed Neural Networks, with RESTful APIs and logging.

• PySpark FIFA Player Clustering
  PySpark ML pipeline for KMeans clustering of FIFA 2018 players by attributes like Overall and Potential. Features preprocessing, feature engineering, and visualization, with FastAPI for real-time predictions.

• Industry 4.0 Predictive Maintenance System
  Production-grade predictive maintenance system using PyTorch CNNs and Siamese networks for real-time hydraulic fault detection, with FastAPI, MLflow tracking, and robust monitoring.

• Counterfeit 3D Component Detection System A comprehensive system that authenticates and detects counterfeit 3D components using advanced pattern recognition and microstructural analysis techniques. This solution is tailored for industries requiring high-fidelity component verification.

• IoT Data Analysis System
  Comprehensive IoT system with ETL pipelines for environmental and traffic sensor data, driving ML-based traffic prediction, real-time processing, visualization, and automated deployment.

• Professional ETL Pipelines (Xetra)
  Python ETL pipeline (Git, CI/CD, Docker) for processing Xetra dataset on AWS S3, enhancing data accessibility.

• Census Data Explorer
  ETL pipeline with SQLAlchemy for U.S. Census data, providing insights via reports and visuals, deployable on Kubernetes.

• MongoDB Nobel Prize Explorer
  Python ETL pipeline on Kubernetes for ingesting Nobel Prize data into MongoDB with NoSQL modeling, improving scalability.

• Advanced Python with OOP Projects
  Collection of object-oriented Python projects, including web apps and CLI tools, showcasing advanced programming.

• Languages: Python, SQL, C++, Shell Script, TeX
• Data Engineering: PySpark, Airflow, DVC, MLflow
• Machine Learning: TensorFlow, PyTorch, Hugging Face Transformers, Scikit-learn
• DevOps: Docker, Kubernetes, GitHub Actions, CI/CD
• Cloud Platforms: AWS, Azure
• Databases: MongoDB, PostgreSQL, MySQL
• Web Frameworks: FastAPI, Flask

• Email: kanhaiya.lgupta21@gmail.com
• LinkedIn: linkedin.com/in/kanhaiya-gupta21/
• Twitter: @kanhiya_gupta21

Explore my repositories and reach out for collaboration or inquiries!