
Ashmi Banerjee
Doctoral Candidate focusing on Recommender Systems & HCI at Technical University of Munich, Google Developer Expert Machine Learning
Munich, Germany
Actions
Ashmi is a doctoral researcher at the TU Munich, focusing on Recommender Systems. She graduated with a master's degree in Computer Science in 2019 and has three years of industry experience in Germany.
A Google Developer Expert and AI/ML Champion Innovator, Ashmi was named one of the 100 technologists to watch for 2023 and has received multiple awards, including the GDE Community Award and the Women Who Code Applaud Her Awards for 2023.
As a Google Women Techmakers Ambassador, she advocates for closing the gender gap in STEM. Ashmi enjoys traveling and training for triathlons. 🏊♀️ 🚴 🏃♀️
Links
Area of Expertise
Topics
SynthTRIPs: Creating Realistic Travel Queries using Gemini for Smarter, Greener Trip Recommendations
Imagine building a travel recommender system that truly understands a traveler’s unique needs — from budget and personal interests to walking distance preferences and even local air quality. Sounds great, right? The challenge is: the datasets we have today don’t capture all those nuances, especially for sustainable travel planning.
In this talk, I’ll introduce SynthTRIPs, a project where we use LLMs (Gemini & Llama over VertexAI) grounded in real-world knowledge about European cities to generate realistic user queries — paired with rich personas and explicit travel constraints. These aren’t just random prompts; they’re fact-checked, diverse, and aligned with people’s real travel goals, including sustainability.
I’ll walk you through how we designed the pipeline, the kinds of personas and queries we created, and what we learned from validating them with both humans and AI tools. Beyond tourism, you’ll see how this approach can improve personalization and benchmarking for recommender systems in many domains.
If you’re interested in personalized AI, recommender systems, or just love the idea of smarter, greener travel tech — this session is for you.
Based on our recently accepted paper: Ashmi Banerjee, Adithi Satish, Fitri Nur Aisyah, Wolfgang Wörndl, and Yashar Deldjoo, in Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’25), July 2025.
[All the data, code & resources are open-sourced for this project]
RAGs to Riches: Generating Sustainable Travel Recommendations with RAG and Gemini
Join us for an engaging session focused on building a Retrieval-Augmented Generation (RAG) system, emphasizing generating sustainable travel recommendations.
Our discussion will center around leveraging Google's Gemini model via Vertex AI to generate text, augmenting this process with relevant content from a vector database populated with Wikivoyage data for 160 European cities.
In addition to enhancing the travel experience through personalized suggestions, we'll introduce Sustainability Augmented Reranking (SAR), a novel approach to integrating sustainability into traditional tourism recommender systems (TRS). This modification incorporates a city's popularity and seasonal demand into the prompt augmentation process, aligning recommendations with sustainability goals. By doing so, the SAR-enhanced RAG system delivers more responsible and eco-conscious travel suggestions, balancing visitor preferences with environmental impact.
Furthermore, 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.
Build your backend using FastAPI
In this talk we will discuss about building the backend for a web-application using Python and FastAPI.
FastAPI is a modern, fast (high-performance), web framework used for building APIs with Python. The session will highlight performing simple CRUD operations using the different endpoints of the API and testing them using pytest.
It will focus not only on the concepts but also emphasise on some of the software engineering best practices (patterns and anti-patterns) required to excel in web-development.
Building a More Inclusive Web: How to Create Fair Recommendation Algorithms
Recommendation systems have become an integral part of the online experience, influencing what we see, read, and buy. However, these systems have been shown to perpetuate and amplify biases, leading to unfair and inequitable outcomes for certain groups of users. In this talk, we will explore the importance of fairness in recommendation systems and discuss strategies for designing and implementing systems that promote equity.
We will examine real-world examples of bias in recommendation systems and consider the ethical implications of these biases. Finally, we will discuss the role that technologists and other stakeholders can play in creating a more equitable web through the design and use of fair recommendation systems
Destroy the dilemma - REST vs gRPC during model serving using TensorFlow Serving
Transitioning from proof-of-concept to deployment is a change many machine learning models struggle to make. While most industrial machine learning (ML) projects target the development of highly performant and scalable ML systems in production, it is often difficult to automate and operationalise these systems so that they work perfectly.
In this talk, we will walk you through the two popular architectures - REST and gRPC using Python, docker and TensorFlow Serving which can be used to deploy your machine learning models in production. We will also discuss the pros and cons of the two architectures, solving your dilemma and ensuring a smooth transition of your models from proof-of-concept to production.
Women Techmakers Conference, Malmö, Sweden #IWD2023 Sessionize Event
WeAreDevelopers World Congress 2022 Sessionize Event

Ashmi Banerjee
Doctoral Candidate focusing on Recommender Systems & HCI at Technical University of Munich, Google Developer Expert Machine Learning
Munich, Germany
Links
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top