FURO · v0.7 · BUILD 2026.04
MATTE BLACK
TEMP · 2700K
N 37° 46′ · W 122° 25′
STRUCTURED MEMORY
Structured memory for agents

Agents are only as good as the memory they can reach for.

FURO turns the things agents read, hear, and write into typed records your team — and your models — can query by field, not by keyword.

Start free See how it works
02 mechanism · how input becomes memory

Raw conversation, typed on arrival.

Every call, email, doc, or tool output lands in FURO and crystallizes into a lattice of typed records — companies, people, deals, parts, orders. No prompt tricks. No per-app glue code.

01 · crystallize

Input arrives as a cloud. It leaves as structure.

Transcripts, tickets, call notes — whatever comes in is parsed into entities, relations, and fields the moment it lands. No re-ingestion, no nightly jobs.

  • entities / record12.4 avg
  • fields / entity8.7 avg
  • ingest latency (p50)240ms
02 · canonicalize

"14.2 meq/100g" and "14.2 milliequivalents" are the same value.

Units normalize. Dates snap to ISO. Entities resolve against your canonical lists. By the time your agent queries, there's one right answer for the same fact.

  • unit taxonomy4,200 canonical forms
  • entity resolverembedding + rules
  • disambiguation acc.99.3%
03 · shared node

One entity. Every memory it appears in.

The same customer, part, or record links across CRM notes, ERP orders, field-service jobs, marketing campaigns. Query the node — get everywhere it's been.

  • link types14 builtin
  • cross-source graphbidirectional
  • dedup strategyconflict-aware merge
03 differentiator · typed recall

Vector search returns relevant-ish chunks.
Typed recall returns the answer.

Below, the same question runs two ways. Vector search surfaces keyword-adjacent noise from unrelated records. FURO dereferences the typed field on the right record — every time, with the right scope.

04 recall · how you ask

Ask a question, not a search.

Typed memory is aggregateable. Ask for rollups, filters, pivots, counts — the same way you'd ask a person who knows your business cold. Get a number, not a list of links.

05 · aggregate

523 deals, one question.

Vector search caps out at "top-k chunks." Structured memory rolls, pivots, and compares across thousands of records — with the same interface a person uses.

  • pivot depthunlimited
  • window functionssupported
  • response time (p95)< 900ms
06 · askable

Ask once. Get the number.

Rolling soil pH? ARR booked this week? Open blockers by stage? FURO returns a value — typed, cited, sourced — not a list of loose links for you to sift.

  • answer typesscalar · table · chart
  • provenanceper-field
  • format guaranteeschema-checked
The substrate doesn't care whether the record is a sales deal, a soil sample, or a repair ticket.
The shape does.
— FURO design principle
05 your business · shaped

Your workspace, as structured data.

Point FURO at your domain and it shapes itself around your vocabulary. The sales team gets deals, attendees, competitors. Agronomy gets blocks, samples, treatments. Same substrate, different schema.

07 · example · sales

A deal card, written by the agent.

A 32-minute call becomes a typed deal card — prospect, stage, close date, owner, blockers, competitors, attendees — all addressable, all queryable, all pivotable.

  • fields extracteddeal: 18 · call: 42
  • entity linkscompany · person · competitor
  • confidence per fieldcited
08 · workspaces

Same substrate. Every business shape.

We don't ship industry SKUs. We ship a memory system. Your schema — tables, fields, relations — is how FURO knows what your business looks like.

  • schema definitionyaml · ui · inferred
  • field types42 builtin · custom ok
  • multi-workspaceshared entities, scoped views
06 shipping path · four steps

From signup to addressable memory, in an afternoon.

No infra to stand up, no vector store to babysit. Connect your sources, describe your shape, and your agents have structured memory the next time they run.

STEP 01

Describe your shape

Define the entities and fields your business runs on. YAML, UI, or inferred from samples.

5 min
STEP 02

Connect your sources

Calls, emails, docs, tool outputs, Slack threads. Ingest flows into the shape.

one-click
STEP 03

Point agents at it

A typed memory tool your agent calls by field. No RAG rewrites, no re-embedding pipeline.

3 lines
STEP 04

Ask, pivot, roll up

People and agents query the same substrate. Answers cite their source — per-field.

live