Francisco Moran Email & Phone Number

Córdoba, Argentina

United States

● Developed and applied RLHF (Reinforcement Learning Human Feedback) strategies to train advanced language models, optimizing chatbot naturalness and effectiveness for industry-leading clients.● Implemented rigorous quality control practices to ensure accuracy and relevance in chatbot responses, enhancing user experience.● Led a quality assurance team, coordinating tasks and setting standards to ensure excellence in AI project delivery.● Supervised and reviewed the work of contributors, ensuring deliverables met the highest standards of accuracy.● Conducted training and onboarding programs for new team members, providing mentorship and technical guidance to ensure effective integration into projects.● Collaborated with cross-functional teams to meet project objectives efficiently and within established timelines.● Facilitated the resolution of complex issues and promoted continuous improvements in quality control processes, driving the delivery of high-quality products across all projects.

United States

• Sentiment Analysis Project on Streaming Service Reviews: -Engineered data by manipulating textual content, conducting pre-processing, and applying TF-IDF vectorization for sentiment analysis. -Generated word embeddings to serve as the foundation for data vectorization. -Trained an SVM-based Sentiment Analysis model proficient in identifying positive and negative sentiments.• Prediction of Vehicle Brand and Model from Unstructured Images in E-commerce: - Visualized, cleaned, and pre-processed a dataset encompassing 196 classes for predicting vehicle brand and model. - Achieved an 82% F1-score by training a classification model using Convolutional Neural Networks (CNN) with the ResNet50 architecture. - Engineered an API service with Flask and incorporated a Redis message broker. Orchestrated services within Docker containers.• Creditworthiness Prediction for Mortgage Applicants: - Managed and visualized a vast dataset of over 350,000 transactions for assessing creditworthiness. - Conducted comprehensive data pre-processing and trained diverse supervised models (RandomForest, XGBoost, LightGBM), yielding a ROC-AUC score surpassing 72%.

Córdoba, Córdoba, Argentina

Córdoba, Córdoba, Argentina

Machine Learning Developer Fellowship

Data Scientist With Python, Information Technology

Ingenieria En Sistemas, Ingeniería De Software

Data Analytics Certificate, Information Technology