Felipe Mello Email & Phone Number

Trenton, Nj, Us

Menlo Park, Ca, Us

Boston, Massachusetts, Us

Mixed SWE/ML research role, focused on NLP. Acted as a tech lead, mentoring a team of 10 engineers. • Significantly improved NLP results by implementing state-of-the-art techniques, Retrieval Augmented Generation (RAG) and creating a data labeling pipeline. Good results led to hiring full time nurses for labeling; • Wrote and maintained a library, used by multiple engineers, that streamlined model development; • Developed and deployed tens of transformers, impacting over 70 services and millions of people;• Pitched and led an initiative that reduced the time to develop/deploy ML models from 10 weeks to few days; • Implemented W&B experiment tracking; • Pitched and led a paradigm change in deployment, from single model instances to multi-modal instances using LoRA adapters, reducing footprint and tech debt; • Currently working on expanding our ML impact by designing scalable modeling processes;

Boston, Massachusetts, Us

Tel-Aviv, Il

Tel-Aviv, Il

Responsible for managing and developing ML/AI solutions. Was promoted to senior in 2023. Supervised a co-op.• Graph Neural Networks: Built a pipeline to process knowledge graphs, train GNNs and interpret their results. Used it to rank and recommend novel associations between biological entities (gene, disease, drugs, etc);• Natural Language Processing: Developed a tool that allowed multiple stakeholders to visualize, label and classify their own data with transformers, so that they could retrieve the information they needed from hundreds of thousands of documents; • Image segmentation: Used OpenCV to label data and CNN to train a deep learning model for nerve density assessment in mice;Responsible for managing and developing ML/AI solutions. Was promoted to senior in 2023. Supervised a co-op.• Graph Neural Networks: Built a pipeline to process knowledge graphs, train GNNs and interpret their results. Used it to rank and recommend novel associations between biological entities (gene, disease, drugs, etc);• Natural Language Processing: Developed a tool that allowed multiple stakeholders to visualize, label and classify their own data with transformers, so that they could retrieve the information they needed from hundreds of thousands of documents; • Image segmentation: Used OpenCV to label data and CNN to train a deep learning model for nerve density assessment in mice;

São Paulo, Sp, Br

• Using python and matlab, I developed a software that utilized smartphone sensors to autonomously map indoor venues and track customers with 2m precision;• Camale was one of 100 startups in the country to be selected for the Brazilian Government acceleration program Inovativa;

Leuven, Be

Worked in the Disruptive Growth Office (ZX Ventures)- Designed and built a tech solution for craft beer recommendations in supermarkets. In storeswith that solution, ABI sold 17% more;- Developed a methodology to efficiently forecast sales results and improve out-of-stock andoverstocking challenges;- Expanded ABI’s craft beer merchandising portfolio by handling everything from productdesign to sales analytics, resulting in a 16% increase in revenue;

New York, Ny, Us

As part of our activities in the Entrepreneurship Center in college, we organized the event TEDxUnespBauru - the first TEDx event of São Paulo State University.

Bauru, São Paulo, Br

Founded and led CEU, an Entrepreneurship Club, composed by over 60 college students that directly impacted more than 1,000 people in its first year through multiple events, workshops and competitions such as Hackathons.

Master Of Science - Ms, Computer Science

Bachelor Of Science In Mechanical Engineering

International Student, Mechanical Engineering