patotricks15

"Interactive Geology: Conscious Society" 1) Database Creation for Mineral Information Led the data science team in establishing a comprehensive database of minerals used on the website, facilitating user access to detailed and accurate mineral information, enhancing educational content and user engagement. 2) Generative AI Directed the implementation of Generative AI technologies (LangChain and Google Vertex AI) for the creation of text, images, and chatbot functionalities on the website. This initiative significantly improved user interaction, provided personalized content, and streamlined user inquiries, setting a new standard for digital engagement in the mineral information sector.

Leadership: Established the data science department within an early-stage startup, fostering collaboration with Business, Product, Marketing, and Software Engineering teams, enhancing communication, ownership, leadership, and collaboration skills. Generative AI: Developed backend systems to create AI Agents from user-submitted documents, refining prompts, and establishing best practices for developing LLM-powered POC/MVP/products using Hugging Face and LangChain. Machine Learning Research: Employed Machine Learning techniques to estimate retention, churn, forecast ROI, and subscription probability using TensorFlow and scikit-learn, delivering information 300% faster with 80% fewer errors than traditional Excel sheets. Applied Technologies: Leveraged LangChain, Hugging Face, Python, Scikit-learn, Numpy, MySQL to develop Generative AI applications and machine learning models, and used Streamlit to build POCs and MVPs.

"Assessment of Poverty in the Brazilian Economy Using Microdata from Continuous PNAD and Machine Learning." Main Field: Applied Social Sciences Field: Economics Subfield: Quantitative Methods in Economics Specialty: Mathematical, Econometric, and Statistical Methods and Models

Exploratory Data Analysis: Conducted exploratory data analysis using Python, pandas, numpy, and Jupyter Notebook to uncover insights, addressing critical business questions and supporting decision-making in real estate financing. Artificial Intelligence: Implemented computer vision using Street View and Google Cloud Vision APIs for advanced pattern recognition, reducing house photo analysis time by 92%. Data Engineering: Created Python scripts for automated document collection and spreadsheet manipulation through web scraping and crawlers, significantly improving operational efficiency and data handling accuracy. Applied Technologies: Utilized Python, Google Cloud Platform, Open Vision, MySQL, Docker, Airflow, requests, and BeautifulSoup for computer vision and ETL pipeline projects.

Product Building: Developed internal and external MVPs, POCs, and tools for geolocated keyword search; created tools for exploratory data analysis focusing on sentiment analysis (NLP) in Play Store comments, generating revenue for the company. Generative AI: Identified bottlenecks in customer service and conceptualized a product using LLM and NLP, reducing costs by about 90%. Exploratory Data Analysis: Produced analyses and reports for the marketing team using statistical analysis, which were published on the company’s website and blogs, increasing visibility. Applied Technologies: Utilized Python, Scikit-learn, Numpy, Matplotlib, NLTK, OpenAI API, GeoPandas, Plotly, Leaflet, MongoDB, Airflow for machine learning models, geospatial analysis, and ETL pipelines in data engineering projects.