Personalization at Agent Scale - Community Meetup

GRAB YOUR SPOT HERE: https://luma.com/xpfemx2v

​Join us for an evening of engineering talks, practical insights, and networking with Vista, OLX, Qdrant, Superoptimal and the AI Collective.

​About The Topic

​Personalization is moving beyond static recommendations and one-size-fits-all search. As AI agents begin to interact with products and services on behalf of users, the infrastructure behind personalized experiences needs to evolve.

​How do you build embeddings that work across an organization? How do you connect customer signals to retrieval and ranking? And how do you serve relevant results at scale, with the speed and reliability that real-time experiences demand?

​We'll bring together engineers from Vista, OLX, and Qdrant to explore the systems powering the next generation of personalization.

​Expect real architectures, engineering trade-offs, lessons from production, and plenty of time to meet fellow builders.

​🗓️ Event Agenda

• ​5:30pm — Doors open | Food, drinks & networking

• ​6:10pm — Vista: From Embeddings to Personalization at Scale

• ​6:35pm — OLX: Building a Unified Embeddings Platform

• ​7:00pm — Qdrant: Retrieval Infrastructure for Real-Time Personalization

• ​7:30pm — Networking, food & drinks

​🧠 Talk 1: From Embeddings to Personalization at Scale — Building Toward Agentic Experiences

​Mouloud Lounaci | Vista - LinkedIn

​Personalization at scale requires more than a great model. It takes a complete stack connecting customer signals, shared embeddings, vector retrieval, rankers, and reliable serving.

​Mouloud will share Vista's journey building fmX, its recommendation foundation model, and the personalization platform around it.

​He'll explore how the team is moving beyond one-off models toward a shared, tiered recommendation stack, including the architecture they're building, the decisions they've made, and the trade-offs they're encountering along the way.

​He'll also look ahead to the capabilities they're exploring next: real-time intent, contextual ranking, and exposing recommendations for agentic experiences.

​This is a journey in progress, with a focus on practical engineering lessons rather than a finished architecture.

​🧩 Talk 2: Building a Unified Embeddings Platform at OLX

​Jose Medeiros | OLX - LinkedIn

​Embeddings are becoming a foundational building block for search, recommendations, and personalization. But what happens when different teams, models, and products all need to use them?

​Building embeddings is one challenge. Turning them into reliable, reusable infrastructure across an organization is another.

​Jose will take us through OLX's work on a unified embeddings platform, exploring the architectural considerations behind making embeddings available across multiple use cases.

​We'll explore the engineering questions that emerge when moving from isolated ML pipelines toward shared infrastructure: consistency, scalability, model evolution, and the relationship between embeddings and downstream applications.

​🔎 Talk 3: Beyond Vector Similarity: Retrieval for Personalized Experiences

​Ewa Szyszka | Qdrant - LinkedIn

​Great embeddings are only part of the equation. To deliver personalized experiences, you need to retrieve the right information, incorporate context, and return results fast enough to keep up with users.

​What does the retrieval layer need to look like when every user, query, and interaction is different?

​Ewa will explore how vector search and retrieval infrastructure can support personalized experiences, from combining semantic similarity with contextual signals to balancing relevance, latency, and scale.

​We'll look at the engineering challenges of connecting embedding models to production retrieval systems, and how those systems can evolve to support more dynamic, agent-driven experiences.

​👋 Who Should Join?

​Whether you're working on AI agents, recommendation systems, search and relevance, embeddings, ML platforms, or vector databases, you'll find plenty to dig into.

​Come for the engineering insights. Stay for the food, drinks, and people. 🍛🍻

​Space is limited — register early.

​Who Are We

​The AI Collective is a global non-profit building the human layer for the AI era. We unite 200,000+ leaders, builders, and stakeholders across 100+ forums worldwide to democratize the frontier, build trust, and coordinate how society navigates the rapid acceleration of technological progress.