CoreCortex Core · developer preview

Memory systems
you can explain.

CoreCortex is an evidence-first engine for configuring, operating, and evaluating retrieval across the storage your corpus actually needs.

CORPUSevidenceversion 24
sourcerecords12,482
strategyhybridcandidate 06
evaluation0.842+ 4.8%
Built for consequential corpora
SUPPORTOPERATIONSRESEARCHENTERPRISE
The CoreCortex approach

Retrieval is not one database call. It is a chain of decisions that should be visible.

CoreCortex gives the corpus a durable contract—what came in, what was derived, where it is indexed, how it is retrieved, and how well that strategy performs.

01

Describe

Version datasets, transformations, indexes, and retrieval policy.

02

Preserve

Keep source identity, lineage, relationships, and access boundaries intact.

03

Retrieve

Return ranked evidence through an explicit, inspectable query plan.

04

Measure

Test relevance, latency, cost, and policy against stable evaluations.

One corpus contract

Keep intent close to implementation.

Datasets, transformations, storage, indexes, and query policies live in a reviewable configuration. Validate it before anything changes.

Read the configuration guide
corec.yamlv1alpha1
apiVersion: corecortex.io/v1alpha1
kind: Corpus
metadata:
  name: support-kb
spec:
  datasets:
    - support-articles
  indexes:
    - lexical
    - vector
  retrievalPlan:
    support-hybrid
Configuration valid · 1 dataset · 2 indexes
Start open. Scale deliberately.

A clear product boundary.

Core stays portable and inspectable. CoreCortex Enterprise operates the pipelines and experiments that become expensive at scale.

OPEN SOURCEDeveloper preview

CoreCortex Core

Build and inspect your own evidence system on open contracts.

  • Declarative corpus configuration
  • Canonical evidence and provenance
  • Normalized record ingestion
  • Pluggable storage interfaces
  • Inspectable retrieval plans
  • Evaluation harnesses
View open-source scope
ENTERPRISEPrivate preview

CoreCortex Enterprise

Continuous corpus operations for production retrieval systems.

  • +Maintained ingestion connectors
  • +Schema and storage discovery
  • +Generated evaluation candidates
  • +Continuous strategy experiments
  • +Promotion gates and monitoring
  • +Enterprise operations
Become a design partner
The planned research loop

Let the corpus prove what works.

Inspired by the disciplined loop behind autoresearch, CoreCortex Enterprise is designed to test one storage, schema, or mapping decision at a time—and keep only measured improvements.

qualityDoes it find the right evidence?
latencyCan it answer on time?
costIs the strategy sustainable?
policyDid it respect every boundary?
Explore CoreCortex Enterprise
EXPERIMENT LOGACTIVE DESIGN
041
baseline · hybrid-v4nDCG@10 · 0.803
BASE
042
section-aware chunksnDCG@10 · 0.831
KEEP
043
graph expansion · depth 2nDCG@10 · 0.824
DROP
044
citation relation boostnDCG@10 · 0.842
BEST
changebuildevaluatedecide
Public documentation

Understand every layer.

Browse all docs
CoreCortex · memory with provenance

Build a system that remembers why.

Start with open contracts today. Help shape the Enterprise ingestion and research layer for tomorrow.

Read the docs →Talk to us