Projects

• •This project aimed to tackle the complexities of media planning across diverse channels, where challenges such as errors and compliance issues frequently arise. These hurdles often result in time-consuming tasks and potential regulatory risks, diverting attention away from strategic thinking and creative endeavours.
• •Our objective was to streamline the media planning process, minimizing errors and ensuring compliance, allowing teams to focus on strategic decision-making and creativity.
• •Built with Angular and .NET Framework on SQL Server, deployed on Microsoft IIS with Firebase for client-side operations, backed by a CI/CD pipeline for reliable, repeatable releases.

What I did

• ›Owned the full project lifecycle on my own: development, deployment, database management and client communication.
• ›Separated the tightly coupled frontend and backend into a modular architecture that was easier to maintain and extend.
• ›Upgraded the legacy .NET Framework 4.8 codebase to .NET Core 7.
• ›Moved 100% of client-side operations from the IIS server to Firebase, and set up the CI/CD pipeline for error-free releases.

• •Developed and deployed a smile detection application for elderly residents in Japanese old age homes using computer vision and machine learning algorithms.
• •Designed and implemented a user-friendly interface using Angular with Flask backend to analyze smiles in real-time.
• •The Python-based detection service ran alongside a SQL Server data layer and was hosted on AWS for elder-care facility access.

What I did

• ›Captured smiles through the webcam and returned real-time feedback to help users improve their facial expressions.
• ›Built a dynamic Angular interface for session management and live display of analysis results and improvement suggestions.

• •The project involves the implementation of a receptionist system for tourists and hotel guests, utilizing a multilingual AI avatar.
• •The system provides assistance and information to guests in multiple languages, enhancing their overall experience while streamlining hotel operations.
• •Built on Node.js and Python, powered by the OpenAI API for conversational responses and served through Nginx.

• •Developed an AI-powered document query system allowing users to upload PDFs and ask questions based on the document content.
• •Designed and integrated an Angular-based frontend with a Flask backend for seamless document processing.
• •Implemented role-based access control, ensuring only authorized users can manage documents.
• •Styled with Tailwind CSS, backed by MongoDB for document metadata, and deployed on AWS.

What I did

• ›Developed and refined the Angular frontend that lets users upload PDFs and query them with AI.
• ›Built UI components for secure authentication, role-based access control and real-time display of AI responses.
• ›Worked closely with the backend and AI teams to bring natural language processing features into the frontend.