Applying LLMs to RecSys: Building a RAG using Google Gemini and MongoDB

Harbour.Space University, 9-11 Carrer de Rosa Sensat, Barcelona, 08005

Explore LLMs in RecSys: integrations, RAG systems, Gemini model via VertexAI, MongoDB augmentation. Deploy with Gradio. Prerequisites: Google Cloud, HuggingFace, Python, ML basics.

May 28, 4:30 – 8:30 PM



Key Themes

AIBuild with AICloudDataGeminiMachine LearningVertex AI

About this event

This workshop will explore the practical applications of Large Language Models (LLMs) in Recommender Systems (RecSys). We'll start with a detailed discussion on integrating LLMs into RecSys research, focusing on different generative paradigms and their pros and cons.

Moving on, we'll provide an overview of RAG (Retrieval-Augmented Generation) paradigms and delve into a hands-on implementation of an RAG system tailored for travel recommendations. RAG serves as an AI framework, grounding large language models (LLMs) with accurate, up-to-date information retrieved from external knowledge bases, enhancing users' understanding of LLMs' generative process.

We will use Google's Gemini model via VertexAI for text generation. Following this, we'll go into the augmentation of our travel queries for the LLM by incorporating relevant content from a vector database created using MongoDB. Our database will leverage Wikipedia data, focusing solely on abstracts, for 160 European cities. Additionally, we'll discover how to deploy our application effortlessly on Gradio—an open-source Python package that simplifies building demos or web applications without requiring any JavaScript, CSS, or web hosting expertise. We also examine the advantages of this approach compared to fine-tuning foundational models and explore methods for evaluating our model's performance.


  • Google Cloud account, GCP credits to access Gemini models via Vertex AI
  • HuggingFace account to deploy our application in HuggingFace spaces
  • Your favorite editor, e.g., VS Code, PyCharm, etc.
  • Proficiency in Python and the basics of Machine Learning
  • Don't forget your laptop to this workshop if you want to try right after

Speaker Bio:

Ashmi is currently a doctoral researcher at the Technical University of Munich. Her research focuses on Recommender Systems and Human-Computer Interaction. She graduated with a master's degree in Computer Science in 2019 from the same university and also holds three years of industry experience at different companies across Germany.

She is passionate about using technology to automate tedious tasks and is always excited to tackle new technical challenges. She has been a Google Developer Expert (GDE) in Machine Learning since 2023.

Ashmi was named one of the 100 technologists to watch for 2023 and won the Google Developer Expert Community Award (Rising Star), the Women Who Code Applaud Her Award (Data Science, 2023), and the DevelopHER Awards 2022 (Emerging Talent).

As a Google Women Techmakers (WTM) Ambassador diversity advocate, she is dedicated to closing the gender gap in STEM through her involvement with various women in STEM networks.

She travels or trains as a triathlete when not sitting in front of her computer. 🏊‍♀️ 🚴 🏃‍♀️



Tuesday, May 28, 2024
4:30 PM – 8:30 PM UTC


  • Ashmi Banerjee

    Technical University of Munich

    Google Developer Expert (GDE) ML, Women Techmakers Ambassador


  • Ekaterina Nosova

    Harbour.Space University

    GDG Organizer BCN, Product Manager


Harbour.Space University logo

Harbour.Space University


  • Joan Fuentes


    Mobile software engineer

  • Carmen Ansio


    UX Engineer

  • Gabriela Soto

    Zurich Insurance

    Data Engineer

  • Rosemarie Garcia


    UX UI Designer

  • Júlia Calderón


  • Javier González Béjar

    GDG Barcelona organizer

  • Ella Quimica

    GDG Barcelona organizer

  • Ekaterina Nosova

    Harbour.Space University

    Product Manager

  • Alex Duran

    Software Engineer

Contact Us