Matthew Meehan Email & Phone Number

Madison, WI, US

Palo Alto, CA, US

Morristown, NJ, US

• Co-led enterprise-wide initiative to integrate LLM-based chat experience into every Majesco product using OpenAI's GPT-4. The chat API enables users to ask any question of their insurance policy, associated claims, and.
• Implemented retrieval-augmented generation (RAG) workflows to enable LLMs to answer questions about larger-than-context unstructured insurance documents.

Morristown, NJ, US

• Developed a product to detect objects and hazards in third-party aerial and streetview images of US addresses to assist underwriting P&C insurance policies. The application consists of 10+ object detection and.
• Trained and deployed photo classification models to detect hazards in photos from property inspection surveys.
• All models were developed end-to-end, from data collection and labeling to deployment and monitoring as containerized microservices on AWS EC2 instances and an ECS cluster.

Madison, Wisconsin, US

Madison, Wisconsin, US

• Back-end and model development for a cloud-based system that uses ML to cleanse semi-structured healthcare provider data, improving provider network accuracy from 50% to >90%
• Data Science lead for proof-of-concept implementations for nations largest healthcare payers (Fortune 50), secured new contracts leading to 8-10x growth en route to Series B
• Involved in the full lifecycle of machine learning projects, from data collection and labeling through model validation
• Created SQL- and Python-based data pipelines for ETL of public and client data sources used to train models

Madison, Wisconsin, US

• Performed large-scale analysis of astrophysical data from NASA’s Fermi satellite on a distributed computing cluster (100 GB - 100 TB scale pipelines)
• Made the first measurement of cosmic-ray proton anisotropy using the Fermi Large Area Telescope with the largest-ever all-sky data set of cosmic-ray protons
• Led data analysis initiatives for citizen science smartphone app that analyzes images to detect cosmic rays
• Developed a convolutional neural network-based computer vision model to identify images of particles taken by a network of smartphones with 95% accuracy
• Worked in an international collaboration of 400+ scientists, coordinated working group, co-chaired weekly international conference calls

Madison, WI, US

• Designed and led two weekly discussion sessions for 50 students enrolled in an electricity and magnetism course
• Responsible for inventing and teaching example problems, designing and grading quizzes, and running separate Q&A sessions for students seeking extra help
• Responsible for running, troubleshooting, and grading one lab session per week, where students were able to explore the concepts from class in a hands-on laboratory environment

Doctor Of Philosophy (Ph.D.), Physics

Master Of Science - Ms, Physics

Bachelor Of Science (B.S.), Physics