Machine Learning Research Engineer
Building efficient, production-ready AI systems that bridge the gap between novel research and real-world impact.

My work revolves around the full lifecycle of machine learning models, from architectural design to deployment and optimization. I'm particularly passionate about:

• Efficient Model Architectures: I am currently working pn increasing the efficiency of Hypencoder models (retrieval models) by reducing the amount of training they require and reducing the size of the model using multi-objective optimization techniques. So far, I have managed to reduce the training duration by 7 times and reduce the computational complexity of the model by 9 times with results comparable to other SOTA retrieval model

• Production-Ready AI Systems: I have engineering and deployed end-to-end ML solutions, for HVAC optimization and scientific information extraction. Most recently I have built and deployed Retrieval-Augmented Generation (RAG) pipeline for a fintech startup.

• MLOps & LLMOps: My interest in AI and machine learning goes beyond model creation and testing into deploying and maintaining models in production. Therefore, I have an interest in the latest MLOps trends and practices to further increase the widespread use of AI.

• Agentic AI: Combing the previous three interests has naturally led me to building AI agents.

Reach out if you would like to discuss anything related to my interests.