I build AI applications and the backend services needed to support them. My technical experience centers on building RAG pipelines and orchestrating multi-agent workflows. I am also actively expanding my cloud skills currently learning how to containerize services and deploy infrastructure on GCP.

• Applied AI Integrations: Building practical applications that connect LLMs to real-world data and user inputs.
• Semantic Retrieval (RAG): Handling document ingestion, text chunking, vector embeddings, and context-grounded response generation.
• Multi-Agent Workflows: Using LangGraph to experiment with multi-step reasoning pipelines that route queries logically.
• Backend Services: Developing FastAPI backends to serve AI models and manage data flow.
• Cloud Exploration: Progressively learning GCP, Terraform, Github Actions and Kubernetes to understand how AI applications are hosted and scaled in the real world.

Developed a multi-agent workflow using LangGraph to handle query intent classification, semantic retrieval, and response generation. Built a document ingestion pipeline with LangChain for text chunking and vector embeddings, integrating a FAISS index for semantic search. Stack: Python, LangGraph, LangChain, Gemini, FAISS, Streamlit

Developed an LLM-as-a-judge evaluation framework using Gemini to assess semantic answer quality against benchmark datasets. Implemented structured output validation via Pydantic to quantify Correctness, Completeness, Relevance, and Hallucination Risk. Stack: Python, Gemini, LangChain, FAISS, Pydantic

Deployed a containerized FastAPI application on Google Cloud Run utilizing Docker. Provisioned cloud infrastructure via Terraform and configured an Nginx reverse proxy. Automated deployments triggered on code commits using GitHub Actions. Stack: FastAPI, Docker, GCP, Terraform, GitHub Actions, Nginx