Senior Data Scientist
Falabella de colombia retail,
Jun 2022 - Current (4 years 1 month)
• Implemented LLM based RAG recommendation system using LangChain, achieving a 13% increase in cross-sell opportunities by suggesting the most likely associated products based on users' browsing and purchase history. • Pioneered the implementation of a two-tower approach for a multitask recommendation algorithm, leading to a 16% boost in sales and revenue. Crafted "Bought together" and "Viewed together" algorithms using this approach. • Engineered session-based recommendation algorithms, including Personalized Trendy and Personalized Popularity, resulting in a 45% increase in revenue and a 20% boost in average order value. • Devised a hybrid search system utilizing semantic and keyword search, incorporating NLP for accurate product rank