Zwe Yaung Ni Tun

I’m an AI Engineer with 2+ years of experience developing AI, machine learning, and full-stack applications for Japanese enterprise clients. I focus on turning agentic AI ideas into dependable systems that work in production. In my current role as an AI Engineer at Global Innovation Consulting (GIC), I build multi-agent systems using LangGraph and design Retrieval-Augmented Generation (RAG) pipelines with embeddings, vector databases, and hybrid search. I also implement LLM evaluation and observability with LangSmith, RAGAS, and DeepEval to improve response quality, faithfulness, and reliability. I’ve deployed AI services on Google Cloud Run and Vertex AI, and I’ve worked with serverless patterns using AWS Lambda, Bedrock, and API Gateway. I monitor latency, token usage, and output behavior so teams can maintain stable production systems, supported by automated testing pipelines. Beyond model quality, I deliver end-to-end experiences: I build frontends with React (and Next.js) and backend services with FastAPI and Spring Boot. I’ve also contributed open-source evaluation work by creating an agent evaluation suite that integrates RAGAS, DeepEval, and LangSmith, and I’ve built applications like a resume & job match platform with 50+ test cases.