Search Technology Meetup - Joint event with MongoDB UG Berlin

We are very excited to announce another Search Tech Berlin Meetup. This time it will be a joint event with the MongoDB Berlin User Group!

Agenda

18:00 - Doors open
19:00 - Introduction
19:10 - First talk
19:50 - Networking, food & drinks
20:10 - Second talk
20:50 - Let’s socialize and grab some drinks

Talks

Building Delivery Hero’s Product Semantic Similarity Using Atlas Vector Search
John Underwood (MongoDB) & Fahad Yousaf (Delivery Hero)

Delivery Hero’s Quick Commerce service provides customers with an easy and efficient way to order grocery items for delivery. However, with thousands of stores and millions of products available on the marketplace, the company faces the risk of revenue loss and customer churn if users struggle to find products that meet their needs and preferences. This talk will explore how Delivery Hero developed a product semantic similarity recommender using transformer-based product embeddings and vector search to identify similar products across various points in the customer’s purchasing journey. We will also discuss the challenges encountered, the solutions implemented, and the next steps for this initiative.

About the speakers

- Fahad is a Machine Learning Engineer at Delivery Hero. My previous roles include working at i2c Inc., and Turing. Over the course of my career, I have developed and productionized a range of machine learning applications, such as Call Transcription Analytics powered by Natural Language Processing, Mobile Remote Deposit Cheque processing using Computer Vision, Semantic Search systems etc.

- John is a Search Specialist with a decade of experience working in the search industry. At MongoDB he works alongside product teams and solution architects helping customers realise their goals for creating groundbreaking applications using the power of search.

From Zero Results to Vector Search: Insights from Nine Key Experiments
Abel Camamacho Guardian & Joel Widmer (Digitec Galaxus)

This session will explore the structured approach we used for experimentation to enhance the user search experience at Digital Galaxus.

We conducted nine key experiments, including four focused on improving the embedding model; one incorporated new signals such as product availability to increase add-to-cart rates, and another optimized latency through vector quantization and the transition from brute-force search to HSNW. Additionally, we examined cases where displaying a zero-results page was more effective than showing vector search results.

About the speakers

- Abel is a Senior Analytics Engineer at Digitec Galaxus with 5 years of experience in E-commerce, working with machine learning to improve search.

- Joel works a Software Engineer at Digitec Galaxus with a master's degree in Mathematics from ETH Zurich. He specializes in developing machine learning models for search.

Many thanks to MongoDB for sponsoring the event and for Delivery Hero for hosting the meetup again!

René & Roman