San Francisco, California, United States
As a data scientist and engineer, I have been involved in multiple projects that leverage natural language processing, unsupervised data mining, and data engineering to generate insights and solutions for various domains. I have worked with the Digital, Data, and Design Institute at Harvard, where I applied NLP and data mining techniques to build Large Language Models tailored for Harvard Business School courses. I have also worked at Systems Limited, where I implemented robotic process automation to streamline and optimize business processes for a large client. Currently, I am pursuing my master's degree in data science at Harvard John A. Paulson School of Engineering and Applied Sciences, where I am also a teaching fellow for the Data Science course. I am passionate about learning and sharing my knowledge with others, and I enjoy mentoring and supporting students in their academic and career development. I am also a cross-registered student at MIT, where I am taking courses on machine learning and data mining to enhance my skills and expand my horizons. My goal is to become a leader and innovator in the field of data science and engineering, and to contribute to the advancement of science and society.
As the Founding Engineer at LaReg AI, I architected a sector-specific LLM, revolutionizing regulatory information processing with a utility-focused chatbot and docket summarization tool. I led the fine-tuning of the 'Mistral' model, utilizing LangChain and LlamaIndex for enhanced accuracy and integrating a knowledge graph for superior contextual insights.
As a Machine Learning Engineering Intern at Vectara Inc in Palo Alto, CA, since July 2023, I've led the development of Vectara-ingest, an impactful open-source repository. It streamlines content crawling and indexing from various platforms like HubSpot, Google Drive, Readthedocs, Docusaurus, and GitHub. My rigorous type checking and integration testing strategies align seamlessly with continuous integration practices, ensuring a smooth, error-free workflow.