Superlinked
Vector compute framework that turns structured and unstructured data into embeddings for RAG and search.
About
Superlinked tackles the unglamorous half of the vector-search stack: not the database, but the compute that turns messy enterprise data into good embeddings. Its Python framework encodes structured fields, text, and images into multi-modal vectors with tunable weighting, so a product's price, stock, description, and image all contribute to retrieval instead of being flattened into one string for an LLM to embed. The company has integration partnerships with MongoDB, Redis, Dataiku, and Starburst.
Founded by Daniel Svonava (ex-Google, YouTube Ads ML infra) and Ben Gutkovich (ex-McKinsey), Superlinked raised a $9.5M seed led by Index Ventures with Theory Ventures participating in March 2024. One caution: the open-source framework repository on GitHub is archived as of April 2026 with 47 stars, and the company now leads with a managed Superlinked Cloud. Evaluate the hosted product, not the repo.
What it can do
- Declarative vector spaces over structured and unstructured data
- Tunable per-field embedding weights
- Real-time embedding updates as data changes
- Semantic search with structured filtering
- Managed cloud with MongoDB and Redis integrations
Setup
- Sign up for Superlinked Cloud at superlinked.com (the OSS framework repo is archived).
- Define your schema: which fields become vector spaces and with what weights.
- Ingest data via the REST API; Superlinked computes embeddings continuously.
- Query with semantic plus structured filters from your app.
Prerequisites
A Superlinked Cloud account; structured and unstructured data to vectorize.
Use cases
Works with
Trust score — 5/10
Scored 5/10 (verified tier, metrics-v1). Evidence: $9.5M seed led by Index Ventures, Theory participating (CityBiz, Mar 2024); founders ex-Google/McKinsey; superlinked.com verified live with MongoDB/Redis partnerships. Strengths: credible team and funding, differentiated vector-compute angle. Docked for: (1) OSS framework repo archived Apr 2026 with only 47 stars; (2) company pivoting to managed cloud, early traction unverified. License: Apache-2.0 (archived framework); cloud proprietary.
★ 47 GitHub stars
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 4.0/10
- Maintenance · recency, releases, issue load
- 4.0/10
- Openness · license + docs
- 5.0/10
- Manual review · sanity read + red-flag check
- 5.0/10
- · $9.5M seed led by Index Ventures, Theory Ventures participating, Mar 2024 (CityBiz)
- · Founders: Daniel Svonava (ex-Google), Ben Gutkovich (ex-McKinsey)
- · OSS framework repo archived Apr 2026 with 47 stars (GitHub API)
- · superlinked.com verified live; MongoDB/Redis/Dataiku/Starburst partnerships
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.