The AI Search Landscape: From Vector Search to Agentic Search
How modern search architectures are evolving to power the next generation of AI applications.
AI search is evolving rapidly. Traditional search engines are adding vector capabilities, while vector databases are expanding into keyword search, filtering, and hybrid retrieval.
But while many platforms now offer similar features, their underlying architectures can differ significantly.
Join AlphaSignal and Vespa.ai for a technical exploration of the evolving AI search landscape and what it means for developers building modern AI applications.
Using a real-world AI shopping application, we'll examine how search requirements evolve from basic vector similarity to sophisticated retrieval systems that combine multimodal data, business signals, machine-learned ranking, and agent-driven search strategies.
What You'll Learn
The Evolution of AI Search: How traditional search engines and vector databases are converging, and why architectural differences still matter.
Beyond Vector Similarity: Combining keyword search, semantic retrieval, structured data, and multimodal signals.
Intelligent Ranking: Incorporating business logic, behavioral signals, and machine learning to deliver more relevant results.
Multi-phase ranking: Understanding how multi-stage retrieval and ranking improve search quality and efficiency.
The Rise of Agentic Search: Enabling AI agents to dynamically select retrieval strategies based on the task.
Where Vespa Fits: Exploring the architectural differences between text-first search engines, vector-first databases, and Vespa.
Who Should Attend?
AI engineers, ML engineers, search engineers, and developers building RAG pipelines, AI agents, recommendation systems, or production AI applications.
Why Attend?
The future of AI search isn't just about finding the closest vector match. It's about building flexible, programmable retrieval systems that adapt to the needs of applications and AI agents.
Learn how modern search architectures are evolving and what it takes to build retrieval systems that go beyond vector search.
About the Speaker
Jenny Morris is a Senior Principal Solutions Architect at Vespa.ai, where she works with engineering teams building production AI search, retrieval-augmented generation, ranking, recommendation, and agentic search systems.
Before joining Vespa, Jenny spent more than six years at Elastic and later worked at Weaviate, giving her a firsthand view of the evolution of modern AI search infrastructure. Her background spans distributed systems, search, databases, and cloud-native infrastructure.
Presented by AlphaSignal × Vespa.ai