This product was not featured by Product Hunt yet. It will not be visible on their landing page and won't be ranked (cannot win product of the day regardless of upvotes).
Product upvotes vs the next 3
Waiting for data. Loading
Product comments vs the next 3
Waiting for data. Loading
Product upvote speed vs the next 3
Waiting for data. Loading
Product upvotes and comments
Waiting for data. Loading
Product vs the next 3
Loading
SemiLayer
The intelligence layer for your existing database
Every AI search tool makes you copy data out and sync a second source of truth forever. SemiLayer flips it: an intelligence layer that bolts onto the database you already run — point at a table, declare a lens, get a typed client. Your data never leaves. Stands out: 4 features in one client (semantic search, similarity, drillable dashboards, personal feeds) where most tools do one; cross-source joins across Postgres/Mongo/ClickHouse; zero vector infra; airgap-capable Runners.
Hey Product Hunt 👋 David here — I built SemiLayer.
Every time I wanted semantic search, "more like this," or a real recommendation feed on my own data, it was the same slog: stand up a vector DB, write an embedding pipeline, babysit drift, keep a second copy of prod in sync forever. The embeddings were never the hard part. The plumbing was.
SemiLayer bolts onto the database you already run — Postgres, Mongo, SQLite, ClickHouse. Point it at a table, declare a lens in a TypeScript config, run semilayer generate:
beam.products.search("eco-friendly water bottle under $30")
Your tables stay in your DB — we query through, we don't replicate them. There's a vector index (semantic search has to land somewhere); we run it so you don't. Need airgap? A Runner in your VPC reads the rows locally and talks out over an outbound-only WebSocket — we never connect to your database directly.
Not just Pinecone with a nicer SDK: four features in one typed client (search, similarity, dashboards, feeds — other tools give you one), cross-source joins across Postgres/Mongo/ClickHouse in one call, ~100–200ms at millions of rows. Free tier is real: 10k rows, no card.
For you if you've got real users on a real database and the smart stuff — search, recommendations, a ranked feed — never made it past "someday." What SemiLayer is for (and, honestly, what it isn't): semilayer.dev/what-is-semilayer
Demo, zero signup: demo.semilayer.com. It's early — what search/rec feature did you quietly give up shipping because the pipeline wasn't worth it? Around all day.
— David
About SemiLayer on Product Hunt
“The intelligence layer for your existing database”
SemiLayer was submitted on Product Hunt and earned 3 upvotes and 1 comments, placing #159 on the daily leaderboard. Every AI search tool makes you copy data out and sync a second source of truth forever. SemiLayer flips it: an intelligence layer that bolts onto the database you already run — point at a table, declare a lens, get a typed client. Your data never leaves. Stands out: 4 features in one client (semantic search, similarity, drillable dashboards, personal feeds) where most tools do one; cross-source joins across Postgres/Mongo/ClickHouse; zero vector infra; airgap-capable Runners.
On the analytics side, SemiLayer competes within Developer Tools, Artificial Intelligence and Database — topics that collectively have 984.3k followers on Product Hunt. The dashboard above tracks how SemiLayer performed against the three products that launched closest to it on the same day.
Who hunted SemiLayer?
SemiLayer was hunted by David Rehmat. A “hunter” on Product Hunt is the community member who submits a product to the platform — uploading the images, the link, and tagging the makers behind it. Hunters typically write the first comment explaining why a product is worth attention, and their followers are notified the moment they post. Around 79% of featured launches on Product Hunt are self-hunted by their makers, but a well-known hunter still acts as a signal of quality to the rest of the community. See the full all-time top hunters leaderboard to discover who is shaping the Product Hunt ecosystem.
For a complete overview of SemiLayer including community comment highlights and product details, visit the product overview.
David Rehmat
Hey Product Hunt 👋 David here — I built SemiLayer.
Every time I wanted semantic search, "more like this," or a real recommendation feed on my own data, it was the same slog: stand up a vector DB, write an embedding pipeline, babysit drift, keep a second copy of prod in sync forever. The embeddings were never the hard part. The plumbing was.
SemiLayer bolts onto the database you already run — Postgres, Mongo, SQLite, ClickHouse. Point it at a table, declare a lens in a TypeScript config, run semilayer generate:
Your tables stay in your DB — we query through, we don't replicate them. There's a vector index (semantic search has to land somewhere); we run it so you don't. Need airgap? A Runner in your VPC reads the rows locally and talks out over an outbound-only WebSocket — we never connect to your database directly.
Not just Pinecone with a nicer SDK: four features in one typed client (search, similarity, dashboards, feeds — other tools give you one), cross-source joins across Postgres/Mongo/ClickHouse in one call, ~100–200ms at millions of rows. Free tier is real: 10k rows, no card.
For you if you've got real users on a real database and the smart stuff — search, recommendations, a ranked feed — never made it past "someday." What SemiLayer is for (and, honestly, what it isn't): semilayer.dev/what-is-semilayer
Demo, zero signup: demo.semilayer.com. It's early — what search/rec feature did you quietly give up shipping because the pipeline wasn't worth it? Around all day.
— David