Projects
Welcome to my projects page! Below you’ll find a collection of selected projects where I’ve utilized my skills in backend development, AI, and cloud computing. These projects showcase my ability to solve complex problems and build scalable, production-ready systems.
Each project is designed to highlight my expertise in Python, FastAPI, AI systems, LLMs, and cloud-native architectures. Feel free to explore the details of these projects and check out the code repositories linked below.
1. Housing Price Prediction Service Link to heading
- Description: This project involved building and deploying a machine learning model to predict housing prices based on a set of features like location, square footage, and number of rooms.
- Problem: The real estate market can be difficult to navigate without data-driven insights into property values.
- Solution: I used regression models to predict prices and deployed the model as a service using Python and Google Cloud Run.
- Technologies Used: Python, Google Cloud Run, ML Models, CI/CD, Docker
- Results: The deployed service enables real-time predictions for housing prices, streamlining decision-making for real estate professionals.
2. Serverless ETL Data Pipeline Link to heading
- Description: A serverless data pipeline to process and transform large datasets for downstream analytics. The pipeline handles data extraction, transformation, and loading (ETL) using Google Cloud Dataflow.
- Problem: Traditional ETL pipelines were slow and resource-heavy, requiring substantial maintenance.
- Solution: I designed a scalable serverless ETL pipeline that efficiently processes large datasets in the cloud.
- Technologies Used: Google Cloud Dataflow, Apache Beam, Python, Cloud Storage
- Results: The pipeline can scale automatically to handle varying data loads, making it easier for businesses to process and analyze data in real-time.
3. Bias Evaluation Analysis for ML Models Link to heading
- Description: In this project, I analyzed structured datasets to evaluate the fairness and bias in machine learning model predictions.
- Problem: AI systems are prone to biases that can lead to unfair or discriminatory decisions.
- Solution: I implemented an analysis framework to measure bias in model predictions and provided actionable insights to mitigate biases.
- Technologies Used: Python, Scikit-learn, Pandas, Bias Mitigation Libraries
- Results: The framework helps improve fairness in model predictions, ensuring that AI systems are ethical and equitable.
(Private research repository)
4. Renewable Energy Financial Modeling Automation Link to heading
- Description: Automated financial modeling for solar power projects using Python and Azure cloud services.
- Problem: The financial modeling process was manually intensive and prone to errors, especially when dealing with large datasets.
- Solution: I developed a set of automated tools to streamline financial modeling, reducing errors and saving significant engineering time.
- Technologies Used: Python, Azure, FastAPI, Data Pipelines
- Results: Reduced manual effort by 70%, allowing faster project assessments and decision-making.
(Internal enterprise project)
5. Real-Time Chatbot with OpenAI API Link to heading
- Description: This project involved building an AI-powered chatbot using the OpenAI API to assist users in a customer support application.
- Problem: Many businesses struggle to provide scalable customer support without a significant increase in resources.
- Solution: I integrated the OpenAI API to create a real-time chatbot that can answer customer queries efficiently and accurately.
- Technologies Used: Python, OpenAI API, FastAPI, Docker, Cloud Deployment
- Results: The chatbot improved customer support response times and reduced the workload on human agents.
How These Projects Reflect My Work Link to heading
Each of these projects demonstrates my ability to solve complex problems through AI systems, cloud-native architectures, and automated workflows. They reflect my focus on delivering scalable and reliable systems that can adapt to real-world business challenges. Whether it’s working with LLMs, optimizing machine learning workflows, or automating data pipelines, my goal is to create solutions that drive value.
Feel free to explore the links to the code to see these projects. If you have any questions or want to collaborate on similar projects, reach out via my Contact me page.