Q

2.0

๐Ÿ›ก๏ธ Data Governance & Quality

Governance is not a layer on top. It is the foundation.

Catalog, quality, privacy, and master data โ€” auto-discovered, AI-enriched, steward-confirmed, continuously enforced. Four products, one semantic layer, one audit trail. SemantIQ catalogs everything. Qualix scores it across 7 ISO dimensions. SenseMask masks the sensitive bits. Entity 360 resolves every identity to one golden record. Governed by Vigil and Zyra.

xAQUA Governance ยท Live
SemantIQ ยท Qualix ยท SenseMask ยท Entity 360
Scanning
โ—Health 91
1,847
Assets
23
Sources
2.4M
Masked
94%
Resolved
๐Ÿ“š SemantIQ
CATALOG
Coverage
0%
customer_360412 assets
hr_master287 assets
finance_gl523 assets
+ 14 moreenriched
โœ… Qualix
7 DIMS
0
Comp
0
Uniq
0
Vald
0
Cons
0
Accy
0
Time
0
Gov
0
Score
๐ŸŽญ SenseMask
LIVE FEED
PII
customer.ssn
Detect
PII
customer.ssn
Partial
PHI
patient.diagnosis
Detect
PHI
patient.diagnosis
Redact
FTI
tax.tin ยท 4 cols
Redact
๐Ÿ‘ค Entity 360
MDM ยท LIVE
CRM
Robert J. Smith
?
BILL
Bob Smith
?
SUPP
R. Smith
?
โœ“ GOLDEN
Robert J. Smith
0.97
โœ“ all-changes-via-CR 14 CRs today 85+ policies audit ยท always-on
The Frankenstack Reality

Five governance tools. Zero of them talk to each other.

Best-of-breed catalog. Best-of-breed quality. Best-of-breed DLP. Best-of-breed MDM. Each works in its own silo. Each tells the auditor a slightly different story.

๐Ÿ“š
The catalog goes stale in 6 months
Traditional catalogs are documentation-first โ€” every business name, every description, every PII tag typed by hand on top of day jobs. Six months in, half is wrong. A year in, half doesn't exist. Documentation-first catalogs scale with human effort โ€” human effort doesn't scale.
27%
typical catalog coverage 11 months in
๐Ÿ“‰
Quality is a weekly script
DQ runs as a weekly script, a monthly audit, or a quarterly ticket. By the time you find the problem, a dozen reports are wrong and three people are in a meeting about it. You find out about bad data after the executive does.
7 days
average gap between drift and detection
๐Ÿšจ
PII leaks happen quietly
Regex in scripts. One-off SQL. A 2019 Python notebook nobody touches. PHI in test. SSNs in a staging bucket. When the auditor shows up, nobody can prove which columns were masked โ€” when, by whom, against which policy. You mask what you remember. The rest leaks.
60%+
of breaches involve mishandled sensitive data
How xAQUA Unifies It

Four products. One semantic layer. One audit trail.

Same catalog. Same lineage. Same policy engine. Same governed change-management workflow. Update a column's business name once โ€” every downstream tool reflects it. Mark a column PII once โ€” masking activates platform-wide. Confirm a join once โ€” every query plan uses it.

01

SemantIQ โ€” the auto-discovered catalog

Connect a source. SemantIQ introspects the schema, drafts business names, classifies sensitivity, scores quality, and surfaces probable joins โ€” 50+ enrichment fields per column, 40+ semantic types, 30+ metrics. Stewards review and confirm via formal change-request workflow. The AI does the heavy lifting. Curation scales. Transcription doesn't.

SemantIQ50+ fieldsZyra (AI Steward)Auto-enrichmentSteward-confirmed
02

Qualix โ€” continuous 7-dimension scoring

Every dataset scored across Completeness, Uniqueness, Validity, Consistency, Accuracy, Timeliness, Governance โ€” continuously. Write rules in plain English (DQO syntax). 83ms average execution on DuckDB. Threshold alerts route to Slack, Teams, or Vigil before the dashboard breaks. SREs treat uptime as continuous. Qualix treats quality the same way.

Qualix7 ISO dimensionsDQO ยท plain English83ms scansContinuous ยท 24/7
03

SenseMask โ€” auto-classify, auto-mask

Detects PII, PHI, FTI, Financial automatically across structured data and documents. 85+ compliance templates pre-mapped to SOC 2, HIPAA, GDPR, CCPA, PCI-DSS, IRS Pub 1075. Three masking methods โ€” Substitute (realistic fakes), Redact (gone), Partial (keep last-4). Referential integrity preserved across tables. Same source value โ†’ same masked value, every time.

SenseMask85+ templatesPII ยท PHI ยท FTI ยท FinancialSub ยท Redact ยท PartialDocs + DBs
04

Entity 360 โ€” one golden record per identity

Resolves "Robert J. Smith" in CRM with "Bob Smith" in billing and "R. Smith" in support โ€” into one governed master record. Identity is a graph, not a row. Continuous matching, merging, and survivorship rules. Stewards review borderline matches via the same change-request workflow. Every xAQUA agent sees the same person.

Entity 360MDM ยท golden recordIdentity graphContinuous resolutionSteward review
How They Connect

Four products. One governed layer.

Auto-discovered by SemantIQ. Scored by Qualix. Masked by SenseMask. Resolved by Entity 360. Orchestrated by Vigil (governance agent) and Zyra (steward agent). Everything reads and writes through the same shared semantic layer โ€” so a change anywhere propagates everywhere, with the audit trail intact.

AI AGENTS ยท ORCHESTRATE GOVERNANCE GOVERNANCE PRODUCTS ยท 4 MODULES ยท ONE SHARED CATALOG SHARED FOUNDATION ยท SEMANTIC LAYER + GOVERNED CHANGE REQUESTS LIVE SOURCES ยท ZERO RAW DATA EVER LEAVES YOUR ENVIRONMENT ๐Ÿง  Zyra ยท AI Data Steward DRAFTS CATALOG ยท CONFIRMS ยท CURATES "Why did customer_profile quality drop?" โ†’ Root cause + suggested fix + CR ready to submit ๐Ÿ›ก๏ธ Vigil ยท AI Data Governance QUALITY ยท PRIVACY ยท MDM ยท AUDIT "Mask all SSN columns across test environments." โ†’ Found 14 cols ยท 2.4M rows masked ยท audit #MK-8824 ๐Ÿ“š SemantIQ DATA CATALOG ยท SEMANTIC LAYER โ†’ Auto-discovered + AI-enriched โ†’ 50+ fields per container/column โ†’ 40+ semantic types classified โ†’ Business glossary ยท cross-catalog โ†’ Lineage graph ยท physical + virtual โ†’ Hybrid search (BM25 + vector) STEWARD-CONFIRMED โœ… Qualix DATA QUALITY ยท 7 DIMENSIONS โ†’ 7 ISO dimensions ยท continuous โ†’ DQO ยท plain-English rules โ†’ 83ms avg ยท DuckDB engine โ†’ Threshold alerts ยท severity tiers โ†’ Lineage-aware root cause โ†’ Project ยท domain ยท enterprise roll-ups CONTINUOUS ๐ŸŽญ SenseMask PII ยท PHI ยท FTI ยท FINANCIAL โ†’ 85+ compliance templates โ†’ SOC 2 ยท HIPAA ยท GDPR ยท PCI ยท 1075 โ†’ 3 methods ยท Sub ยท Redact ยท Partial โ†’ Referential integrity preserved โ†’ Docs + DBs ยท PDF ยท Word ยท CSV โ†’ Auto-classify ยท human-in-loop review AUDIT-COVERED 100% ๐Ÿ‘ค Entity 360 IDENTITY ยท MASTER DATA ยท MDM โ†’ Golden record per identity โ†’ Identity is a graph, not a row โ†’ Continuous match ยท merge ยท survive โ†’ Cross-system across N sources โ†’ Survivorship rules ยท governed โ†’ Steward review ยท borderline matches ONE TRUTH ๐Ÿง  SemantIQ ยท Shared Semantic Layer + Governed Change-Request Workflow One governed catalog under strict change-management. No one edits directly โ€” every change flows through a CR, reviewed by owner/steward, accepted or rejected with reasoning logged. ๐Ÿ“– Business Glossary "Customer" means what your team means ๐Ÿ•ธ๏ธ Lineage Graph Physical + virtual ยท click-through to source ๐Ÿ›‚ Change-Request Workflow Audit-grade ยท owner/steward approval ๐Ÿ“œ Audit Trail ยท Full Operation Log Who ยท what ยท when ยท how many ยท which policy live read ยท schema metadata only ยท zero raw data ever leaves โ„๏ธ Snowflake โšก Databricks ๐Ÿ˜ Postgres ๐Ÿ—„๏ธ Oracle โ˜๏ธ Salesforce ๐Ÿ“„ Files ยท S3 ๐Ÿ“‹ SAP + more N sources โœ“ Same semantic layer across all 4 products ยท โœ“ Every change governed by CR workflow ยท โœ“ Audit-grade ยท always on ยท โœ“ Zero raw data ever leaves
Governance products ยท 4 modulesShared semantic layer + governed CR workflowAgents ยท Vigil + ZyraLive sources ยท zero data movement
Qualix ยท Seven Dimensions ยท One Score

Every dataset, continuously scored.

Trends tracked. Thresholds enforced. Alerts before drift hits your dashboard. SREs treat uptime as continuous โ€” Qualix treats quality the same way.

91
Customer Domain ยท Health Score
Across 23 tables ยท 847 columns ยท 1.5K+ records/scan ยท DuckDB engine ยท 83ms average
โ–ฒ +2 this week
94
โ–ฒ +1 ยท 7d
Completeness
Are required fields populated? Nulls, empties, placeholder values.
97
โ€” stable
Uniqueness
Are primary keys unique? Duplicates surfaced by scan, not by production.
92
โ–ฒ +2 ยท 7d
Validity
Does the value match format, range, and domain rules? Email shapes, date bounds, enums.
78
โ–ผ โˆ’4 ยท 7d
Consistency
Does the same fact agree across systems? Cross-source reconciliation. โš  upstream drift
89
โ–ฒ +1 ยท 7d
Accuracy
Does the value match ground truth? Reference data, lookup tables, external validators.
95
โ€” stable
Timeliness
Did the data land on time? Freshness SLAs, staleness detection, late-arriving rows.
88
โ–ฒ +3 ยท 7d
Governance
Ownership, classification, and documentation coverage across the catalog.
# Written by steward ยท compiled by Qualix
rule customer_id_is_unique
  on    "sales.customer"
  assert "customer_id" is unique
  and    "customer_id" is not null
  severity critical
  threshold 99.5
  alert   "#data-stewards"

โ”€โ”€ last run โ”€โ”€
  scanned   1,547,203 rows
  unique    1,547,198  (99.9997%)   โœ“ PASS
  nulls     0                          โœ“ PASS
  duration  71ms

โ”€โ”€ trend ยท last 7 days โ”€โ”€
  mon  99.9998   thu  99.9998
  tue  99.9997   fri  99.9997
  wed  99.9997   sat  99.9997
                      sun  99.9997

STATUS: GREEN ยท next run in 23 min

Write rules in plain English. DQO compiles them.

Data Quality Objects let stewards write rules the way they'd describe them โ€” "customer_id must be unique and non-null" โ€” and Qualix compiles them into real, tested, observable assertions. No notebooks. No cron jobs. No brittle Python scripts.

  • Rule-based assertions on any column or combination
  • Severity tiers, thresholds, and routing โ€” per rule
  • Reusable templates for common domains (customers, transactions, claims, accounts)
  • Full audit trail โ€” who wrote it, when it ran, what it scored
  • 83ms average execution on DuckDB โ€” no Spark cluster to warm up
SenseMask ยท Detect ยท Classify ยท Mask

Sensitive data. Hidden. Forever.

Auto-detects every type of sensitive data โ€” PII, PHI, FTI, Financial โ€” and applies the right masking policy. Three methods: Substitute (realistic fakes), Redact (gone), Partial (keep last-4). Referential integrity preserved across tables.

Column ยท Source
Before ยท raw
After ยท masked
customer.ssn
PII ยท SSN
123-45-6789
โ†’
Partial
XXX-XX-6789
patient.diagnosis
PHI ยท ICD-10
E11.9 ยท Type 2 diabetes
โ†’
Redact
[REDACTED]
payment.card_number
FIN ยท PCI
4532 1098 7654 3210
โ†’
Partial
**** **** **** 3210
tax_records.tin
FTI ยท IRS 1075
84-1234567
โ†’
Redact
[FTI ยท NOT EXPORTED]
customer.name
PII ยท Name
Sarah Chen
โ†’
Substitute
Martha Ellis
85+ Compliance Templates ยท Out of the Box

Pick a framework. Get a policy.

Pre-built mask sets aligned to the regulations your auditors care about. Ship compliant from day one โ€” editable per tenant, version-controlled policy library.

SOC 2
Service organization controls. Security and confidentiality principles.
PII masking ยท audit evidence ยท access logs
HIPAA
US health information. Safe-harbor de-identification for analytics.
PHI redaction ยท 18 identifiers ยท BAA-aligned
GDPR
EU data protection. Subject rights, pseudonymization, minimization.
DSR support ยท pseudonymization ยท erasure
CCPA ยท CPRA
California privacy rights. Sensitive personal information handling.
SPI classification ยท opt-out ยท deletion
PCI-DSS
Payment card data. Cardholder data masking and tokenization.
PAN masking ยท scope reduction ยท tokenize
IRS Pub 1075
Federal Tax Information. Stringent handling for tax authorities.
FTI classification ยท access logs ยท audit
GLBA
US financial institutions. Non-public personal information.
NPI detection ยท masking ยท safeguards
FERPA
US student education records. Identifier and record protection.
student PII ยท grades ยท transcripts
Entity 360 ยท MDM ยท Golden Record

Identity is a graph. Not a row.

Three fragments. Three systems. One person. Entity 360 resolves them into a single governed golden record โ€” continuously, with steward review on borderline matches and full survivorship rules.

CRM
Robert J. Smith
DOB 1962-03-14 ยท SSN ***-**-7821
rjsmith@example.com
BILLING
Bob Smith
DOB 1962-03-14 ยท SSN ***-**-7821
123 Main St ยท Acct #44219
SUPPORT
R. Smith
DOB 1962-03-14
+1 (555) 482-7821
โค
RESOLVE
โค
โœ“ GOLDEN RECORD ยท CERTIFIED
Robert J. Smith
entity_id: ent_7a3f9c2e ยท match_score: 0.97
DOB
1962-03-14
SSN
***-**-7821
PRIMARY EMAIL
rjsmith@example.com
PHONE
+1 (555) 482-7821
ADDRESS
123 Main St
LINKED SOURCES
CRM ยท Billing ยท Support
🤝
xAQUA augments your governance team, not replaces it. Stewards stop typing descriptions by hand and start curating AI-drafted ones. CDOs stop chasing five tools for one audit answer. Privacy officers stop writing regex โ€” they set policy, and the platform enforces it. Stewards stay in charge. The AI does the heavy lifting.
4 โ†’ 1
Governance Tools
Catalog ยท Quality ยท Privacy ยท MDM unified
83ms
Quality Scan
Continuous ยท 1.5K+ records/scan
85+
Compliance Templates
SOC 2 ยท HIPAA ยท GDPR ยท PCI ยท IRS 1075
Weeks โ†’ Seconds
Audit Response
Lineage and policy on every asset
Customer Story ยท In Production
A $300B+ public pension fund consolidated 4 governance tools into one engine โ€” in an air-gapped deployment.
The catalog lived in one tool, quality rules in another, PII detection in a third, master data in a spreadsheet on someone's desk. Every audit cycle was a three-week scramble. xAQUA consolidated all four onto a single engine โ€” SemantIQ for the catalog, Qualix for the 7-dimension quality scores, SenseMask for HIPAA + IRS Pub 1075 masking, Entity 360 for member-identity resolution across eight legacy systems. Auditors now self-serve in seconds, with cryptographic lineage on every record and a complete change-request audit trail.
4 โ†’ 1
Governance tools consolidated
8 systems
Unified to one golden record (Entity 360)
8ร—
First-year ROI
Ready to start?

Stop bolting governance on top.
Make it the foundation.

See SemantIQ, Qualix, SenseMask, and Entity 360 running together on your data โ€” auto-discovered catalog, 7-dimension quality scoring, 85+ masking templates, golden-record MDM โ€” in a 30-minute demo.