• Developed high-quality Jupyter Notebooks in Python to support the Conversational AI Chat system for one of the top 5 Big Tech companies in the USA (FAANG). • Built and optimized Retrieval-Augmented Generation (RAG) systems to improve model efficiency and accuracy, ensuring adherence to internal documentation. • Trained the AI model to leverage the company’s APIs in conjunction with its own knowledge, enabling the generation of richer, more informed responses. • Created Supervised Fine Tuning (SFT) datasets to teach the model the correct approach aligned with company guidelines. • Conducted comparative analysis across different Conversational AI systems to identify improvement areas within our own platform. • Analyzed response outcomes to detect performance patterns, utilizing reinforcement learning algorithms to enhance positive behaviors and penalize suboptimal ones. • Reviewed and proposed improvements for Python code generated by the AI model, ensuring efficiency and correctness. • Performed data analysis, machine learning, and statistical analysis to assess the model’s decision-making and suggested corrective actions when necessary. • Identified and corrected code errors generated by the AI, ensuring quality and reliability in outputs.

• Consolidated 50 disparate metrics from 13 countries and 5 business units into a single source of information for key performance indicators (KPIs) and customer experience (CX) goals. • Creation of supervised and unsupervised models to support decision-making and prescriptive analytics. • Utilized Big Data techniques, including developing HIVE queries, to meet stakeholder data requirements. • Implemented a Semi-Supervisioned Machine Learning model to detect the cost of each contact for the whole business (For each Business Line and Regions). From this, the business stakeholders were able to allocate the resources more effectively, knowing beforehand how much each contact would cost to the business. • Automated tasks using Python and various APIs to streamline processes and allocate time effectively. • Leveraged business intelligence (BI) techniques to deliver actionable insights and trends on business KPIs to stakeholders. • Collaborated with Product Areas to enhance and develop features with data-driven support.

• Spearheaded task automation, business report generation, and ETL queries to facilitate end-to-end support for the data analysis process. • Creation of supervised and unsupervised models to support decision-making and prescriptive analytics. • Created an unsupervised Machine Learning model to detect a city similar to Bogota, from where it was ran an A/B test and we could make sure the feature implementation was depleting the results of CPO (Contacts per Order). • Developed predictive models to enhance data analysis and yield deeper insights. • Provided global support to squads on internal initiatives related to data analysis and automation within the Global Data Analytics domain. • Collaborated with Product Areas to enhance feature development with data-driven support. • Recognized multiple times for contributions, achieving an A/S performance rating in 2022, primarily for efficiency enhancements and discoveries benefiting the operations team through shared data. Allowing, through all conducted analyses and the introduction of Machine Learning models, a reduction in CX operational costs by 1 million dollars per quarter

• Applied Python for data analysis and statistical tests, particularly in CX Value Generation, generating business insights to enable better resource allocation with a focus on CSAT, NPS, and refunds. • Created a Semi-Supervised machine learning model called Judgement Model to detect whether or not a client were supposed to be given a refund. • Developed an Unsupervised Machine Learning Model to cluster three kind of drivers: Gold, Silver and Bronze, enabling the business stakeholders to know which kind of users to put more focus and effort in. • Creation of supervised and unsupervised models to support decision-making and prescriptive analytics. • Designed automations to enhance the scalability and ease of data analysis processes. • Developed a Predictive Model to accurately describe CPO (contacts per order), surpassing previous estimations by the business. • Enhanced spreadsheets to facilitate stakeholder analysis and modified queries in Hive (SQL-HQL) to retrieve required data efficiently.

• Pioneered the development of FP&A Dashboards using Tableau, providing stakeholders with previously inaccessible or complex data visualization. • Leveraged Excel, VBA, and Power Query extensively for data processing and spreadsheet automation. • Implemented Supervised Machine Learning models to predict the capacity required in the Call Centers. • Provided support in budget planning and financial control for the CX Team. • Conducted financial analysis of CX KPIs to identify opportunities for expense optimization. • Implemented a new Workflow to capture financial data from Focal Points across three business lines and 15 countries, aiming to initiate forecasting and enhance business efficiency. • Assisted in the implementation of a new financial model with the team.

• Leveraged Power Query, Excel, VBA and Python extensively for the meticulous processing and analysis of laboratory data, with a specific focus on discerning the kinetic parameters associated with Fibrous Sorghum.• Applied Time Series analysis in order to understand how the mass was varying across the time.• Published a comprehensive research article detailing the findings and insights derived from the analysis, contributing valuable knowledge to the field of Sorghum research and its potential applications.