WonderSearch

WonderSearch is a search product from Evokoa that lets people upload documents and query them in natural language, without a separate embedding pipeline. It returns passages with source references, and can be used directly in a workspace or built into an application, help center, or AI assistant.

What It Is

WonderSearch describes itself as a serverless sparse retrieval solution for semantic search. Users upload files (PDF, TXT, MD), ask questions in their own words, and receive relevant passages from those files. The site states there are no upfront ingestion or embedding fees, and that search compute is charged per query with storage billed separately.

How You Use It

WonderSearch can be used in two ways: inside a workspace for personal research, team knowledge, and AI assistants, or embedded in a product for in-app search, self-service support, and AI features. Developers can connect through the REST API, Python SDK, CLI, or MCP. Each search can be run at a Small, Medium, or Large effort setting, giving harder questions more room to explore, though the site notes more effort does not guarantee a better result.

Product Context

The site includes pages for benchmarks, comparison against dense embeddings, and documentation. Evokoa describes WonderSearch as backed by Y Combinator, and the team page lists Dale Everett Ng as CEO and Dalton Prescott Ng as CTO.