Adelaide, South Australia · Open to AI, ML and data roles in Adelaide or remote
I build AI agents and retrieval augmented generation systems that go from prototype to production. I'm an AI Developer Intern at ExpertEase AI in Adelaide and a Master of Data Science (Artificial Intelligence) student at Flinders University, where I hold the Go Beyond Excellence Scholarship.
- Now: AI Developer Intern at ExpertEase AI, working on AI agent development
- Before: two years as a freelance AI engineer shipping RAG chatbots, autonomous agent systems and AI automation across 4+ client deployments
- Enterprise background: Associate Engineer at L&T Technology Services, building Azure cloud applications and SQL Server data layers
- Research: book chapter with Cambridge Scholars Publishing and a paper presented at ICATIA 24
- Certified: Microsoft Azure AI Engineer Associate (AI 102) and Azure AI Fundamentals (AI 900)
- Looking for: internship, graduate, casual, part time and contract roles in AI engineering, machine learning and data science
A visa aware job search assistant for international students in Australia. SEEK has no visa filter, so a LangGraph pipeline searches SEEK, reads each full job ad, checks it against the user's visa (500, 485, 482 or PR), ranks jobs on visa fit and resume skill match, and generates tailored resumes and cover letters. Exposed as a FastAPI REST API, an A2A server and an MCP server, with LangSmith tracing, Docker, Kubernetes manifests and Cloud Run deployment.
Python LangGraph Playwright Groq FastAPI MCP A2A LangSmith Docker Kubernetes Cloud Run
An autonomous project building agent. Give it an idea and a tech stack, and a LangGraph workflow plans the work, generates a runnable starter app, packages it as a zip and can push it straight to GitHub. FastAPI backend, Streamlit interface, Docker Compose, Kubernetes and a GitHub Actions workflow publishing images to GHCR.
LangGraph FastAPI Streamlit PyGithub Docker Kubernetes GitHub Actions
Ask questions about any YouTube video in plain English. A four stage RAG pipeline with Google Gemini embeddings and FAISS gives under 500 ms semantic search across transcripts of 100K+ tokens.
LangChain Gemini FAISS Streamlit Docker Kubernetes
A minimal, working Agent to Agent (A2A) workflow with the A2A Python SDK: a Greeting Agent server publishes its Agent Card and a client discovers and calls it.
Python A2A Starlette Uvicorn uv
- Book chapter: A Real Time Vehicle Counting System for Detecting Traffic Surveillance, in Advancing Engineering and Technology through Interdisciplinary Approaches, Cambridge Scholars Publishing, 2026
- Conference: ICATIA 24, International Conference on Advanced Technologies and Intelligent Automation, May 2024
- Microsoft Certified: Azure AI Engineer Associate (AI 102), Aug 2025
- Microsoft Certified: Azure AI Fundamentals (AI 900), Aug 2025
- Python Programming, Infosys, Sep 2023
- Machine Learning, E Cell IIT Hyderabad, 2023
Hiring for AI, ML or data in Adelaide or remote? Let's talk: devachaudhari22@gmail.com · Portfolio