Collaborating with data architecture, data governance, and security teams to ensure solutions and meet required standards in the Solutions Architect Design document (SAD) for deploying Healthcare Recommender Systems in production, covering: • Content recommender • Collaborative filtering recommender • FAQs • Classes and Programs recommenderWorking with data scientists to design effective workflows and architectures for activating Recommender System ML models, optimizing impact in real-time streaming and offline batch scenarios.Implementing data processing workflows, incorporating necessary data cleansing and imputation logic.Addressing technical debt in ML infrastructure, focusing on enhancements to data pipelines, ML compute, data storage, and the Recommender system API, utilizing technologies like Azure Data Factory, Azure Machine Learning, Databricks, AKS, Cosmos DB, FASTAPI, and APIgee.Collaborating with the data engineering team to ensure timely delivery of data in the required format and cadence for ML model training.Experimenting with Large Language Models (LLM) models to effectively summarize and recommend healthcare articles.Member of the MLops council to set the MLops strategy and best practices for Kaiser Permanente.

Stabilized and enhanced Azure Data Factory critical pipelines to reduce cost, time, redundancies, and complexity.Developed custom orchestration process to trigger over 800 ETL jobs in Azure Data Factory to coordinate data cross 9 KP regions. High complexity in the solution that includes token activation, dependencies validation, schedulers (daily, weekly, and monthly), error handling, notifications, and process recovery.

Architectural Runway - TeamExplored, tested, and developed POCs of latest Azure releases to provide new capabilities and increase efficiency to extract, transform, and load data (Azure Data Factory) from a wide variety of data sources using SQL (Synapse and Hive).Supported and collaborated with business users and value streams to recognize and help adopt best practices in ETL pipelines and database promotions.Designed, implemented, and supported automation python scripts to promote objects from Azure Data Factory and Synapse from lower to higher environments for CI/CD. Simplifying self-service support for value streams (Visual Studio 2019, SSDT, Azure Data Studio, Python, SQL, Github, Jenkins, DevOps)Refactored four python frameworks for the NexGen Accounting Program (NCAP). Changed functional programming style to object-oriented programming, wrote automated tests using a mix of unit and Behavior-driven development.File transfer automation:Developer lead for the Plan Partner Vendor Management Framework. This solution allows internal teams at Kaiser Permanente to manage vendor relationships organized by project and automate the transfer of files for internal processes.Framework designed and implemented using Python3.8 (Flask, Paramiko, Apscheduler, PyTest, Git, vscode, mkdocs) database storage in Oracle schemas, Webapp running on IIS in a Windows server.

Developed and deployed on production the Transfer Files Automation framework. A web app to automate and optimize files transfer between servers. Using Agile methodology through Jira. Tools and languages used: Python, Django, Apscheduler, Watchdog, Paramiko, Pipenv, Git, PyTest, Oracle DB. Algorithms include job scheduler, files transfer, emailer, password manager.

Lead developer for the computer vision module in Gap, an open source ML data engineering framework. Duties include design, coding, testing, and progress reporting. Tools and languages used: Python, Keras, openCV, HDF5, Git, PyTest. Algorithms include image transformations, image augmentations, dataset management, persistent storage, streaming, feeders for neural networks.

## Portland, OR Mar-May 2018 ##Kaggle CompetitionTF-NW Team • Participated at iMaterialist Furniture Competition as a member of the TensorFlowNW team (GitHub Repo) ending at the 73 position in the private leaderboard of 428 teams.• Tested several visual recognition models of Machine Learning such as RESNet, Inception, NASnet_Large, and NASnet_mobile. Using Keras and Tensorflow, explored data augmentation, TTA, and different platforms such as Google Colab and Intel DevCloud to train and deploy the models.## Portland, OR Jun-Jul 2018 ##OSCON 2018 – TensorFlow day• Helped Manceps team prepare for the 2018 OSCON Convention.• Demonstrated with TensorFlow for Poets on Google Colab how to download a data set, train, freeze, compile and deploy the model on a Raspberry Pi vision kit to identify objects in real time with its camera.## Portland, OR Mar 2018 - Present ##TensorFlowNW-meetup • Helped as an Event Organizer in the monthly TensorFlow meetup.• Presented: Animated Neural Networks, Kaggle results, Open Source

Freelance: Coached Customer Service Manager on database best practices• Designed Extract, Transform, and Load (ETL) pipeline from raw information until final dashboard• Designed database structure for three businesses of Mexichem (Mexichem Resins, Mexichem Compounds, Mexichem Derivatives)• Linked tables from a variety of sources (SAP, MS Excel, and CSV) to MS SharePoint allowing real time information flow and version control- Built relational databases using MS Power Query in MS Excel- Created online dashboard views using pivot tables, pivot charts, and slicers as decision-making tools for managers• Made the following recommendations to management:- Upgrade MS Excel dashboards to MS Power BI or Tableau- Switch to SQL Server (or similar infrastructure) to improve data security, stability, and the management and storage of large datasets- Build flow charts to detect redundancies in processes and bottlenecks

Responsible for market strategies and database management• Built and administered physical data and digital databases (housed in SQL Server) of personal information and ensured our company followed Habeas Data law• Managed data entry team to digitalize printed forms to the database and physical data storage• Automated and analyzed metrics in Google Analytics through an API connection (built in python) from elespectador.com to develop strategies for editors and writers to improve content and traffic• Analyzed subscriber profiles with the use of information hosted in the CRM (housed in SQL Server) software crossed with TGI, EGM studies, and geo-positioning to develop strategies for increasing subscriptions• Automated data compilation and transformation for daily dashboards and reports using Python, Power Query, and Stored Procedures in SQL Server• Administered and developed strategies for mass communication using mailing campaigns

TV advertising planner in charge of 6 clients and responsible for managing media databases (TV, Radio, printed media, and websites)• Analyzed business information from Caracol Media Group customers to develop their marketing strategies• Created dashboards and reports allowing company executives to evaluate and market programming based on common KPIs• Designed and administrated MS SharePoint structure for the team to centralize and control version for documents, presentations, reports, calendars and dashboards• Administrated MS SharePoint users to share information with other teams and external customers• Utilized MS Excel and Access to optimize marketing dashboards and reports allowing for efficient daily, weekly, and monthly updating• Cleansed customer account data for team’s databases• Developed inventory control database in MS Access to track company merchandise

Responsible for developing the annual survey and present result to managers of Mexichem Resins, Mexichem Compounds, and Mexichem Derivatives• Designed and conducted the annual customer satisfaction survey• Analyzed survey results, identified shortcomings in internal procedures and developed corrective action and strategies to enhance customer relations• Created an application in MS Access to receive and track customer complaints and suggestions

• Designed and implemented the ScoreCard Balance project, providing an effective way to present monthly progress reports to directors and managers.• Utilized MS Excel to generated dynamic reports of the management indicators and results.• Created database in MS Access to efficiently administer information regarding employee work performance.