SCIKIQ · Adaptive Enterprise Intelligence Fabric
From business context to actionable intelligence
UnderstandConnectContextualizeReasonAutomateLearn
Connect the enterprise. Understand its business. Give AI context. Let agents act, safely.
0Model
Industry Knowledge & Reference Layer
External knowledge that seeds enterprise context before a single table is read.Learn this layer →Industry reference models (APQC, BPMN, SCOR)Reference architectures per industryBest practicesStandard ontologies (IFRS, SNOMED, ISO)Taxonomies and standards (GS1, UNSPSC, NAICS)Regulations and compliance (ESG, SOX, HIPAA)Public data and benchmarksLLM knowledge (general and domain)Change intelligence (news and regulation watch)Framework and ontology importers (APQC, SCOR, OWL)Jurisdiction packs (regulators and regulations per country)
1Model
Business Context Construction
Industry-aligned enterprise context, built from LLMs and the enterprise's own knowledge.Learn this layer →Inputs: industry references, process maps, documents, workshops and SMEsEnterprise profile: holdco, opcos, legal entities, segments, lines of business, geographyRegulations bound to entity, line of business and jurisdictionCompetitive position and SWOTStakeholders on an influence and interest gridPersonas and responsibilitiesData domains and sub-domainsDomain 360: owner, steward and everything per domain
2Model
Logical Enterprise Model
The industry value chain and process hierarchy, mapped to data domains, personas, decisions and actions.Learn this layer →Industry value chainL1 value chainsL2 processesL3 sub-processesL4 activities and use casesMapping to data domainsMapping to personasMapping to decisionsMapping to governed actions
3Model
Semantic & Ontology Construction
Taxonomy, glossary, concepts, logical entities and relationships: one shared meaning.Learn this layer →Taxonomy (hierarchy)Glossary: terms and business definitionsConcepts (typed semantic contract)Logical entities (business objects)RelationshipsBusiness rulesGoverned ontology editor (maker-checker)Rules engine and duplicate reconciliationEvidence fabric: provenance on every object
4Ground
Enterprise Discovery Engine
The top-down logical model meets bottom-up metadata discovery.Learn this layer →Top-down: industry, process, domain, persona, concept, metric, taxonomy, glossaryOntology reconciliationBottom-up: applications, databases, APIs, BI, documents, IoTLogical systems and flows (landscape)Technical metadata (types, keys, nulls, ranges)Business metadataAutomatic classificationLineage, including SQL lineage from views and proceduresUsage patternsOwner, criticality, AI suitability and allowed actionsSchema-change rediscovery, process mining and domain induction
5Ground
Logical → Physical Mapping
Map logical entities to physical sources, virtually, with no data movement required.Learn this layer →Logical customer to SAP, Salesforce, Oracle, Snowflake and Power BIConfidence score and source evidenceMeasured data qualityOwnershipMeasured refresh, latency and usageLogical metrics realised in several systems
6Context
Enterprise Knowledge Graph
One graph of business, data, analytical and operational knowledge.Learn this layer →Graph layers: industry, value chain, process, domain, entity, concept, metric, dashboard, decision, actionPersonas, stakeholders, systems, documents, rules, policies, lineage and eventsPersisted graph with traversal and path API, agent tool, Neo4j and OWL exportRecord-level business events (created and updated, with keys)
7Context
Semantic Data Layer
Business objects and virtual semantic models; physical storage becomes an implementation detail.Learn this layer →Business objectsSemantic models and virtual tablesMetrics and KPI modelsLogical relationshipsREST and GraphQL endpointsMulti-hop traversal, multi-object query and natural language to SQL
8Context
Connectors: SAP, Oracle, Salesforce, Snowflake, SQL, APIs, files, BI, IoT and streamingQuery and API federationPushdown optimisationCachingOn-demand materialisationIncremental ingest (upsert and merge)Data quality rulesTransformation (derived attributes)CDC and streamingMaster data and entity resolutionCertified data productsData marketplace and data exchangeQuery planner, tier advisor and refresh scheduling
9Context
Metrics & Knowledge Fabric
Parse BI reports, dashboards and semantic models into a governed metrics layer.Learn this layer →BI discovery and metadata (Power BI, Tableau, Looker, Qlik, Excel, dbt)Metric extraction: formula and dimensionsBusiness meaning and definitionsMapping to domain and processKPI to decision to actionDocuments, events, policies and rulesOperational knowledge: hybrid keyword and vector searchKPI definitions: target, thresholds, directionDimensions and metric historyWhy did it change? Driver decompositionMetric implementations reconciled across systemsBI lineage to metrics (Power Query, DAX, visuals)BI report usage: used, unused and file-fed
10Context
Persona, domain, process, entity and metric contextHistorical context and memoryPolicy contextCurrent state: live KPIsAvailable actionsContext packs: relevant data, rules, metrics, policies and permissionsContext pack generator per persona, served by API and versioned per runSubject 360 and predictions in context
11Act
Agent builderAgent registry and domain agentsAgent runtime with tracingOrchestration and multi-agent systemsShort- and long-term memoryTools and planning: tool registry (HTTP, OpenAPI, MCP, plugins)Agent templates: Sales, Finance, Operations, Supply chain, Customer service, RiskStreaming and asynchronous runsHuman-in-the-loop: pause, ask, approve, resumeAgent-to-agent interoperability (agent cards, remote agents)
12Act
Decision & Reasoning Engine
Use context, rules and models to determine what should happen.Learn this layer →Retrieve contextReason with LLMs and modelsCalculate and simulateEvaluate rules, policies and constraintsOptions and a recommendationPredictive models: train and score
13Act
Identity and access (RBAC and ABAC)Information classification: public, internal, confidential, PIIData security and privacy: PII, encryption, maskingPolicy and compliance (SOX, GDPR): scope, condition, severity, strictest winsConfidence and explainabilitySegregation of dutiesAudit and lineageTool-call guardrails: schema, budget, side-effect approvalApprovals: human path for sensitive actions, automatic within policyRow-level ABAC, consent and purpose, single sign-on
Approved actions → Action Fabric
14Act
APIs, integrations and RPAWorkflow automationTransactions and data updatesNotifications (email, Teams, Slack)Task managementTarget systems: SAP, Salesforce, ServiceNow, ERP, CRM, email, Teams and custom APIs
15Learn
Outcome & Observability
Measure outcomes and feed continuous learning back into every layer.Learn this layer →Action outcomeBusiness outcome: KPI change against a baselineAgent performance and prediction accuracyModel performance: tokens, latency, costEvaluation harness: judges, golden sets, regressionTracing (OpenTelemetry)AdoptionExceptionsHuman feedbackContinuous learning: outcomes improve ontology, metadata, context and agents
Continuous learning loop: outcomes improve ontology · metadata · context · agents
SCIKIQ — Connect the enterprise. Understand its business. Give AI context. Let agents act, safely.
Discover → Model → Ground → Contextualize → Reason → Act → Learn
Learn every layer
The Data Academy teaches the fabric one layer at a time: concepts, what good looks like and the questions to ask.