I’m a Software Engineering student at McGill University with a strong interest in backend development and applied AI systems.
I enjoy building reliable AI solutions that combine LLMs, RAG retrieval and backend engineering to solve real business problems.
Currently, I’m working as a Backend AI Developer Intern at Ciena, where I focus on improving the reliability and evaluation of multi-agent RAG systems used for cloud emulation portal.
- Python, Java, TypeScript
- FastAPI (async), REST APIs
- Git, Docker
- Large Language Models (LLMs)
- Retrieval-Augmented Generation (RAG)
- Multi-agent systems & agent orchestration
- Prompt engineering & tool calling
- LLM evaluation (LLM-as-a-Judge)
- Pandas, NumPy
- Data preprocessing & analysis
- Performance metrics & benchmarking
- Azure / OpenAI APIs
- Jupyter Notebook
Tech: Python, FastAPI, LLMs, RAG, async programming
- Designed and improved a multi-agent RAG pipeline for internal documentation search
- Reduced answer inconsistency using agent memory and noise filtering
- Implemented agent orchestration (preprocessing, tool calling, sub-agent routing, post-processing)
- Built automated evaluation using LLM-as-a-Judge and performance metrics
➡️ Private enterprise project — architecture and learnings documented
- Built a RAG-powered enterprise chatbot using LangChain and Dify, integrating with a MySQL database to retrieve customer business data and reduce manual workload for client success and strategy teams.
- Designed an AI workflow automation system using LLMs to generate image prompts and marketing strategy prompts from keywords and visual inputs, accelerating content creation for marketing teams.
Tech: Java, Spring Boot, MVC, REST APIs
- Collaborated in a 5-member team to build a fitness center management platform using a layered MVC architecture with Spring Boot.
- Applied dependency injection and inversion of control to decouple components, improving maintainability and extensibility.
- Designed RESTful APIs with structured exception handling and request/response DTOs to ensure clean frontend–backend communication.
Tech: Node.js, Prisma, PostgreSQL, REST APIs
- Collaborated in a 7-member agile team following Scrum methodology with 2-week sprints and daily stand-ups; served as Scrum Master to coordinate progress.
- Architected backend infrastructure using Node.js and Prisma ORM with a PostgreSQL database, implementing secure RESTful APIs for financial data management.
- Ensured ACID compliance for all financial transactions and supported real-time budget calculations.