Jorge Guerra

Onkopilot

San Juan, PR

• •Co-founded AI-powered oncology support platform for cancer patients and caregivers
• •Built the marketing website and currently developing the web application using a RAG (Retrieval-Augmented Generation) system for personalized oncology guidance
• •Leading technical architecture leveraging LLMs and medical knowledge bases for accurate, empathetic patient support

Xsporty

San Juan, PR

• •Co-founded sports platform connecting athletes to find games and build community
• •Developed the full product stack: web app, iOS app, Android app, and all backend services
• •Architected and built real-time game matching, user networking, and scheduling systems from the ground up

Popular Bank

San Juan, PR

• •Co-designed, implemented, and managed a cutting-edge training program leveraging Microsoft Power Platform and Python to enhance workforce expertise in process automation, data analytics, and data science
• •Led Data Analytics and Data Science Programs, overseeing end-to-end analytics workflows — defining success criteria, generating insights, and delivering actionable recommendations
• •Managed performance metrics, aligned cross-functional stakeholders, and developed interactive dashboards and reports for data-driven decision-making
• •Designed scalable analytics solutions, built robust data models, and implemented machine learning models to optimize business operations and drive efficiency
• •Fostered a data-driven culture, standardizing methodologies to assess project value, process improvements, and strategic investments across teams
• •Unclaimed Property — Saved $58M by developing a tool to assess customer behaviors and classify claimed vs unclaimed properties data to accurately provide government clients correct information
• •Loan Portfolio Valuation — Saved $300K annually by implementing advanced data processing and optimization techniques for portfolio asset valuation, forecasting, and amortization
• •Customer Lifetime Value Prediction — Developed ML models to optimize marketing campaigns, predicting short- and long-term customer value across diverse segments
• •ATH Mobile Alert Automation — Built an ML model for AML alert classification, distinguishing between SAR and No SAR, reducing analyst workload by 10X

MYH Solutions LLC

Dallas, TX

• •Designed, developed, and managed the universal order system for Virtual Care Coordinators (VCC) implemented in outpatient neurology clinics
• •Leveraged automation and implemented data entry improvements to analyze, report, and enhance the clinics' operational digital processes
• •Streamlined workflows and increased efficiency, empowering Virtual Care Coordinators to provide more effective and timely services to patients

My Dental Path

San Juan, PR

• •Led all technical development for a platform empowering internationally trained dentists to navigate US dental school admissions
• •Built and maintained the website including service pages, bundled package flows, and content management
• •Managed the full technology stack from front-end design to hosting and deployment

One Drop

Remote

• •Performed research and development of a cardiovascular diseases risk model, augmenting a well-known categorical risk score to include continuous measurements such as blood glucose, blood pressure, BMI, weight variability, and activity calories to estimate CV risk score and its variability over time for people with diabetes
• •Worked on ranking of diabetes-related measurements to find and measure their importance and association between different physiological events
• •Designed and developed an insights system to provide useful, timely, and accurate information about different day patterns to improve communication and services between coaches and customers

KPMG US

Dallas-Fort Worth, TX

• •Developed a process automation tool to obtain, analyze, and provide real-time insight into the audit of medical records using software development, computer vision, and NLP, automating tasks while improving time, accuracy, and consistency of record management processes
• •Developed and implemented a computer vision model to detect and process assets' tags to improve workflow in warehouse management

Children's Hospital of Philadelphia

Philadelphia, PA

• •Technical lead of a group (two staff members and four students) which designed, developed, and deployed a recommendation system analyzing CHOP and U-Penn researchers' publications and staff skill sets to provide meaningful collaborative recommendations — currently in production
• •Technical lead of a clinical research group which designed and developed an algorithm to predict the likelihood of patients missing their appointments — won The Drexel LeBow Analytics 50 Award
• •Developed and deployed automation scripts to support ETL of data from IBM Netezza CHOP Data Warehouse prior to machine learning model analysis
• •Served as liaison between the Advanced Analytics Department and hospital staff requesting data analysis, data modeling, and information visualization

Columbia University

New York, NY

• •Conducted research in human genetics, developing a model to provide patient information based on variants from exome sequencing data using read depth information
• •Investigated the value of imputation from exome sequencing to capture lower frequency variants in isolated populations to identify mutations that may increase the risk of common diseases
• •Developed quality control tools to preprocess data and run different methods of burden test to classify association to specific traits

Columbia University

New York, NY

• •Performed research in learning and classification of IMU sensor data using logistic regression, Hidden Markov Models and other ML techniques
• •Developed a system to accurately predict and quantify the type and accuracy of upper body primitive motions of stroke patients
• •Published: Capture, Learning, and Classification of Upper Extremity Movement Primitives, IEEE ICORR 2017, London

Intel Corporation

King of Prussia, PA

• •Developed, executed, and debugged compute fabric validation plans and tests for Intel HPC products
• •Employed lean/agile practices for product software and system release quality
• •Developed programs to provide reliable and consistent test infrastructure for network systems and components

Intel Corporation

Folsom, CA

• •Designed, developed, and validated testability circuit software for Non-Volatile Memory
• •Developed and debugged complex software programs to convert design validation vectors and drive test equipment
• •Tested, validated, modified, and re-designed circuit test programs to guarantee component margin to specification with emphasis on yield analysis

UCF Intelligent Systems Lab

Orlando, FL

• •Conducted research in context-based learning and collaborative context-based reasoning applied to the development of a multi-agent collaboration system

UC Berkeley

Berkeley, CA

• •Conducted research in probabilistic inference and sequential decision making under uncertainty in robotics applications

Intel Corporation

Folsom, CA

• •Worked on development and debug of wafer level parametric, functional, and characterization tests, focusing on automation and analysis tools in NAND Memory Technology products
• •Gathered and analyzed data from wafer sort production runs and engineering experiments to guarantee component specifications and performance
• •Gave a Perl training class to Probe Engineers regarding methodologies to develop probability plots and automated detection of run-to-run yield toggles

Intel Corporation

Folsom, CA

• •Process validation and knowledge of general test methodologies during the Ivy Bridge project
• •Built, debugged, compiled, and tested development tools in Visual Studio C# .NET and Perl
• •Code development and automation of manual and functionality test cases
• •Integrated, debugged, validated, and released module/test programs for HVM customers

Boeing

Bellevue, WA

• •Implemented automation on the CAD/CAE/KBE tool CATIA V5
• •Provided support for knowledge-based related products using the .NET programming language