I have over 14 years of experience as a specialist in Artificial Intelligence, Machine Learning Engineering, Data Engineering and Software Engineering. During this time, I have developed complex models and applications for a range of industries, including the healthcare domain, decision-making systems in emergency response, recommendation systems, and Internet of Things (IoT) environments.

Featured Projects

• Healthcare Multi-Source Agentic RAG Platform
  LangGraph, Amazon Bedrock, OpenSearch, AWS Glue, Athena, Postgres, Redis, FastAPI, Docker
  A healthcare question-answering platform for multi-source documents and structured data, using LangGraph routing, agentic RAG, GraphRAG, vector search, conversation memory, and cloud-backed ingestion workflows.
• Healthcare Multimodal Assistant
  Gemini, Vertex AI, Imagen, FastAPI, Docker, GCP
  A full-stack healthcare assistant app using Google Vertex AI (Gemini), Imagen, FastAPI, and Docker for multimodal chat, document/image/video analysis, and API integration.
• Multimodal AI Application
  HuggingFace, Gradio, PyTorch, Whisper, BLIP, SAM, NLLB, BART, VITS
  A Gradio-based web app providing speech recognition, image captioning, segmentation, translation, summarization, and TTS using HuggingFace Transformers and PyTorch.
• Multimodal PDF Parser & Abstract Generator
  Gemini, Mistral, PyPDF2, FastAPI, Redis, Docker
  A PDF parser and abstract generator using Gemini for AI summarization, Mistral for OCR, and FastAPI/Redis for scalable job queueing.
• Automated Medical Case Review System with LangGraph
  LangGraph, OpenAI LLMs, FastAPI, Docker, Python, CSV, workflow automation
  An automated system that reviews medical cases from CSV uploads, providing summaries, risk assessments, and recommendations. Uses LangGraph for agent-based workflow orchestration, supports Docker deployment, and features a web interface for results.
• Book Recommendations with RAG + Airflow
  Airflow, RAG, Weaviate, FastAPI, Ollama, Docker
  An end-to-end book recommendation system using Retrieval-Augmented Generation, Apache Airflow orchestration, Weaviate vector search, and a FastAPI web interface. Supports ingestion, semantic search, and chatbot recommendations for books, orchestrated by Airflow DAGs and powered by local LLMs and embeddings.
• Full-Stack Azure LLM RAG Application
  Azure OpenAI, RAG, FastAPI, Docker, AI Search, Web App, CI/CD
  A full-stack Azure OpenAI assistant with RAG, combining FastAPI backend, static frontend, Azure AI Search, and containerized deployment. Features chat and RAG modes, vector search, and CI/CD with GitHub Actions.
• Control Point Deformation Network
  CPDNet, Medical Imaging, 3D Spine, Deep Learning, Python
  A deep learning framework for patient-specific prediction of postoperative 3D spine shape in Adolescent Idiopathic Scoliosis (AIS) surgery, combining geometric accuracy and anatomical plausibility for clinical decision support.
• Multi-Agent Banking Workflows with CrewAI
  CrewAI, FastAPI, Docker, Python
  A workflow-first multi-agent banking system using CrewAI for customer support, sales, and content pipeline automation.
• Designing a Langchain Chatbot with RAG
  LangChain, RAG, FastAPI, Python, Docker
  A LangChain-powered chatbot with RAG, FastAPI backend, and live-data services for weather, news, and web search.

See all projects →

Skills & Technologies

• LLMs (Gemini, GPT-4, Llama, Mistral)
• Multimodal AI
• Computer Vision
• NLP
• CrewAI
• LangChain
• RAG
• MLOps
• Python
• PyTorch
• TensorFlow
• Docker
• Kubernetes
• Cloud: GCP, AWS, Azure