Generative Ai, Machine Learning Engineer
Current> Developed a personalized Goal template recommendation system using Collaborative Filtering and Matrix Factorization to enhance user engagement, resulting in a 25% increase in click-through rate (CTR).> Fine-tuned OpenAI API parameters to create dynamic recommendations of mentors and goal templates to mentees with high relevance by 90%.> Led the design and deployment of a content-based recommender system for an online mentor-mentee platform, enhancing mentors and mentees experience by generating personalized goals, books, courses and events suggestions, which improved goal completion rates by 20%.> Collaborated with data scientists and engineers to optimize Transformer-based models (BERT, GPT-3 and GPT-4) for text generation tasks, reducing model inference time by 30% while maintaining high-quality outputs.> Built and deployed a Neural Collaborative Filtering (NCF) model for recommending products based on user-item interaction data, achieving a 50% lift in purchase conversion rates.> Integrated a Retrieval-Augmented Generation (RAG) system, using OpenAI API, Llama, Hugging Face and LangChain for document retrieval and GPT for response generation, improving the relevance of long-form recommendations by 20%.> Experimented with Variational Autoencoders (VAEs) for generating new content-basedrecommendations, leading to 12% higher recommendation novelty.> Incorporated a Reinforcement Learning algorithm to dynamically adjust recommendations in real-time, improving user interaction time by 18%.> Integrated a natural language processing (NLP) pipeline to analyze customer reviews and fine-tune recommendations based on sentiment.
