A short glossary
What we do, in plain words
Seven ideas behind Flowlexi, PaveDB, and the book Inspectable Retrieval with PaveDB — explained for people who run businesses, not servers. Each one links to where it lives in our stack.
Semantic search
Classic search finds the words you typed. Semantic search finds what you meant: ask for “contract cancellation deadlines” and it also surfaces the clause that says “either party may terminate within 30 days”, even though no word matches.
It works because text is compared by meaning, not spelling. For a business this changes what a knowledge base is worth: people find answers phrased differently from their question — which is most answers.
PaveDB is a search engine built this way — pavedb.org Back to top ↑
Embeddings
An embedding model reads a piece of text and places it as a point on a very large map, where distance means similarity of meaning. “Refund policy” and “money-back rules” land close together; “refund policy” and “parking rules” land far apart.
Semantic search is then just: place the question on the same map, look at what sits nearby. Which model drew the map — and which version of it — decides everything about the results.
An inspectable vector database
A vector database stores those map points and finds neighbors fast, across millions of documents. Most treat this as a black-box index: results come out, and that is all you can say about them.
An inspectable one keeps the receipts: every stored item knows its source document and the exact model that placed it on the map; every query can be replayed later and produce the same answer. When a result feeds a decision someone signs, the receipts are the point.
The idea has a book — inspectableretrieval.com Back to top ↑
RAG — retrieval-augmented generation
Language models are fluent but they answer from memory. RAG makes the model look things up first: retrieve the relevant documents, then write the answer from them, with citations.
The quality ceiling of a RAG system is its retrieval. If you cannot audit what was retrieved and why, you cannot explain the answer — which is why we start from the search side.
Portable, declarative AI flows
Most AI automations are glue code: logic buried in scripts only their author can follow. A declarative flow states the steps in a readable file — retrieve, decide, act — like a recipe rather than a trail of pans.
Readable files can be reviewed, versioned, and moved between environments. The automation stops being a person’s secret and becomes a company asset.
Self-hosted models
The models that read and write text — encoders and decoders — can run on hardware you control instead of a third-party API. Your documents never leave; costs are a machine, not a meter; behavior doesn’t change because a provider shipped an update.
The trade-off is honest: you operate the machine. That is exactly the part we sell as a service when you’d rather not.
Managed instances, your keys — cloud.flowlexi.com Back to top ↑
The cloud console
Everything above, running in minutes without hiring infrastructure people: the console provisions your search instance, manages keys and billing, and keeps versions upgraded.
You keep the keys, the data, and the exit: the engine is open source, and your archive leaves with you whenever you want.
The fastest way to see all of this working is a running instance.
Start free on Flowlexi Cloud