Pooja Baralu Umesh

Machine Learning Engineer · NLP, Retrieval, MLOps

Hi, I’m Pooja, a Machine Learning Engineer passionate about turning data into impactful, production-ready AI solutions. With a Master’s in Data Science and hands-on experience fine-tuning diffusion models, building recommendation systems, and orchestrating MLOps pipelines, I love tackling challenges at every stage of the ML lifecycle—from data collection and feature engineering to model deployment and monitoring.

Over the past year, I’ve:

  • Designed and deployed a Retrieval-Augmented Generation system for wine product search using vector embeddings, OpenAI LLMs, and Pinecone.
  • Fine-tuned Stable Diffusion with LoRA for domain-specific image generation and evaluated results with CLIP similarity.
  • Built scalable ML APIs with FastAPI, Docker, and Kubernetes, integrating models into real-time applications.

My core interests include deep learning, generative AI, system design, and MLOps. When I’m not coding, I’m reading fictions, watching reality shows, or exploring the latest advancements in AI research.

Feel free to explore my projects and blog posts, or connect with me on LinkedIn and GitHub!