Herman

Data Scientist

University of Florida•  August 2022 - August 2024•  Gainesville, FL, USA

Organize data processing scripts into a custom, in-house, proprietary python package, DRAPI-Lemur. Use shell scripting in BASH to parallelize data processing. Create command-line Python scripts using the argparse package. Use GitHub actions for Continuous Integration / Continuous Delivery (CI/CD). Optimize SQL database ETL using Pandas and SQLAlchemy custom chunking methods resulting in 100% fewer server-side interruptions due to memory-constraints. Write and develop extraction and transformation scripts for data transfers from IDR to researchers. Provide statistical summaries and insights for data released to research parties. Suggest and apply, when possible, technical improvements such as debugging logs to automatic pipelines to more easily identify problems. Follow Python programming best practices according to PEP conventions, such as formatting, naming conventions, coding style, programming integrity, etc. Document ETL scripts with README files. Specialize in datasets sourced from Epic electronic health records (EHR) and the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). Communicate clearly with coworkers on task blocks to identify problems and solution in daily SCRUM meetings. Participate in team education meetings by providing different viewpoints that might help the team. Contribute to organizational meetings by setting agendas, providing meeting space, stimulating discussion, prompting for input.

Data Scientist

State of Florida•  May 2022 - August 2022•  Tallahassee, FL, USA

Provide statistical advice on fraud detection strategies as Systems Project Consultant. Write software in Python to collect information for fraud prevention using third-party APIs. Interpret and suggest improvement on Tableau visualizations. Learn laws and rules regarding US and Florida health care.

Customer Service Representative

Florida Department of Management Services•  December 2021 - May 2022•  Tallahassee, FL, USA

Rated as personable and patient by QA Coach. Helped employees of the state of Florida with questions about their retirement.

Data Scientist

Florida State University•  January 2020 - December 2021•  Tallahassee, FL, USA

Used GPT-3 and EleutherNeo for knowledge extraction of unstructured data in Natural Language Processing prototyping. Scraped 40+ unique websites with Selenium for Python and Web Scraper for Nutrition Facts information in text and image formats. Converted image formats of Nutrition Facts into text using Tesseract for Python (Pytesseract). Parsed 2050 extracted Nutrition Facts strings into each nutrition component using general expressions. Wrote a visualization function for locating keywords and their associated terms with arrow markers and color highlights. Found cliques among nutrition components using the Networkx implementation of the Bron-Kerbosch algorithm. Compared text similarity using a Python implementation of gestalt pattern matching (Ratcliff/Obershelp Pattern Recognition). Submitted a SBIR/STTR project pitch that resulted in an invitation for a full proposal submission.

Data Science Lecturer

Florida State University•  January 2019 - May 2021•  Tallahassee, FL, USA

Performed term-frequency analysis using machine learning on students' self-introduction as a possible way to identify special needs students. Diction did not significantly correlate with class grade. Wrote Python script to grade nearly 100 online homework assignments from KhanAcademy for each of 60 students. Graphed distribution of grades to identify students with special needs revealing a bimodal distribution among 60 students. Analyzed data from instant feedback technology (iClicker) to identify weak spots in students' knowledge. Investigated and implemented different educational technologies to solve specific learning goals. KhanAcademy for student homework, Kahoot for engaging class activities, Canvas for grade management, and iClicker for collecting real-time data on students’ performance. Developed own slides, homework, and assessments tailored to local student's needs. Taught one section of an introductory class on statistics with 60 students.

Chemist

Miami-Dade County•  January 2011 - March 2015•  Miami, FL, USA

Designed experiments to validate current standard operating procedures which led to a statistically significant increase in accuracy. Researched literature to develop methods for storing and neutralizing waste locally. Performed quality assurance and quality control (QA/QC) on chemical analysis data. Interacted with clients and provided technical assistance and materials. Wrote and edited standard operating procedures (SOP). Served as Safety Officer and Co-Technical Director.