Document models & feeding
We design the schemas that describe your content and build the pipelines that feed documents into Vespa and keep them up to date.
Vespa.ai consulting
We build new applications on Vespa, improve existing deployments and migrate from Elasticsearch or OpenSearch, handling the data pipelines, ranking and integration.
What we work on
Retrieval & ranking
We combine keyword and vector retrieval, filters and ranking signals around your content. We evaluate results against the queries and tasks your application needs to support.
Application integration
We connect Vespa to your software through query APIs and data pipelines, so retrieval and ranking work within the product. For .NET applications, we maintain Vespa.NET.
Migration to Vespa
We assess whether Vespa fits your Elasticsearch or OpenSearch application, adapt document models, queries and integrations, and validate search quality and performance against your data. We plan the rollout around your production requirements.
Vespa.NET
We maintain Vespa.NET, an open-source .NET SDK for Vespa. Define document models, deploy schemas, feed documents and query from your application.
dotnet add package Vespa.NETSearch Match for Vespa
Find which saved YQL searches match an indexed document.
Indexed documentTrail shoeCategory: shoesPrice: €89
Saved searches
- Shoes under €100✓ Match
- Jackets under €100No match
Validation and handover
How we workWe agree how to evaluate search quality using your data and representative queries, then test throughout the build. Response times, operating costs and monitoring are part of the engineering work. Production handover includes the code and architecture documentation your team needs to continue.
Let's talk about your Vespa project
Tell us about your data, your application and the results you need from Vespa.