Luis Patricio Moreno Buitrón

I was the data architect for a large bank and acted as the counterpart of the international consultant for the creation and startup of the BI unit in that financial institution. I was in charge of the implementation of the IBM data model for financial data warehouses, the definition of the bank's BI architecture, the establishment of the operating model of the BI unit, and the definition of the catalog of BI services that would be offered to the business areas of the bank.

I was part of the team of enterprise architects who generated a model that conceptually and technically supports the implementation of the client's transformation strategy in terms of business and technology.

The project's specific objectives were to:
-Identify improvements at the level of business processes, governance, information systems, procedures, and current regulations, which must be considered for a new implementation of the applications that support the company.
-Define the new data architecture with which information management can be improved.
-Identify the new architectural plan of the applications that support the company and from which the transition strategies of the current applications can be defined.
-Define a new deployment architecture for applications in accordance with the technological infrastructure of the client.
-Identify the current and future situation of the architecture, determine the gap, and establish the respective roadmap.

A Data Mart and ETL solution for the consolidation of data about application monitoring taken from the APM tool database of a company. The solution encompasses four components: Extraction, transformation, load, and utility components.

An application to mask the data of the databases of a USA company according to the PII regulation. The application has two components: A programmatic component to execute a PII scanning of the databases to detect PII data and a programmatic component to complete the masking of the PII data detected.

A solution framework to obtain metrics about the performance of the AWS Autora database instances and generate performance reports. The framework has the following component: A repository for the collected metrics, a component to connect to the database instances to collect the metrics defined, a component to generate the diagnostics based on the variations of the metrics collected, and a component that will generate performance reports.

Oracle database migration between data centers and six clinics that are part of a healthcare and insurance company. The databases in question are classified as critical, so the defined migration strategy was based on Oracle MAA. It was oriented to minimize downtime, with which it was possible to migrate each database in an average time of 15 minutes.

Designed and developed an Applied AI MVP for a university IT service desk to improve ticket intake, classification, routing, and SLA-risk visibility. The solution combined LLM-based text enhancement with traditional machine learning models to transform noisy user-generated ticket descriptions into cleaner, more reliable inputs for downstream prediction.

The MVP included a working application that allowed support staff to enter or review ticket information, improve ambiguous titles, classify requests into the appropriate category or support queue, and identify tickets at higher risk of SLA non-compliance. Historical service desk data was profiled, cleaned, anonymized, and used to train and evaluate the models.

The project covered the full lifecycle from data preparation and feature engineering to model training, evaluation, application integration, and reporting. Performance was assessed using metrics such as precision, recall, F1-score, confusion matrices, and ROC-related analysis where applicable. The solution was designed as a practical prototype to demonstrate how AI could improve ticket quality, routing consistency, prioritization, and early operational decision-making.