Blog

Insights & Updates

Thoughts on metadata engineering, data mapping, and building audit‑ready data infrastructure.

October 9, 20267 min read

The Technical Debt Magellan Retires

Not all technical debt lives in application code. Here is the inventory of data-transformation debt Magellan eliminates outright — and how each one is retired by design.

October 8, 20266 min read

Does Magellan Answer Every Data Governance Question?

An honest map of what Magellan fully answers, what it answers thinly, and what it leaves to the rest of the governance stack — because coverage claims that ignore gaps do not survive a procurement review.

October 8, 20267 min read

How Magellan Provides Data Lineage and Provenance

Trace source-to-target field relationships, preserve versioned mapping decisions, and verify their history with Magellan’s lineage and provenance architecture.

October 8, 20268 min read

Transformation Debt, and the Record That Pays It Off

Transformation logic that lives only in scripts and memories compounds quietly. Here is how Magellan turns it into versioned, governed, reproducible decisions — and makes the whole history tamper-evident.

May 6, 20268 min read

How Magellan Uses Ontologies

Ontologies give Magellan a governed concept layer above source and target schemas — so mappings resolve to shared meaning, not just matching strings. Here is how that layer works, end to end.

May 5, 20269 min read

AI Agents Need a Metadata Layer — Why Autonomous Systems Can't Operate Without Governed Data Context

AI agents are only as reliable as the data they access. Without a metadata layer that provides schema context, lineage, and semantic meaning, agents hallucinate, break pipelines, and make dangerous assumptions.

May 3, 20268 min read

Data Catalogs vs. Knowledge Graphs — Why Passive Inventory Isn't Enough

Traditional data catalogs inventory your assets but can't reason about them. Knowledge graphs capture relationships, context, and lineage — turning metadata from a filing system into an intelligence layer.

May 1, 20269 min read

Data Quality Engineering — Shifting Left to Catch Bad Data Before It Breaks Everything

Stop treating data quality as an afterthought. Learn how to embed validation, profiling, and anomaly detection directly into your pipelines — and why metadata-driven governance is the key.

April 30, 20268 min read

CMMC L2 §3.3 AU-Family Controls, Satisfied by Construction — One Hash Chain, Three Regimes

How Magellan's Audit & Trust Ledger produces the evidence a CMMC L2 §3.3 C3PAO assessor needs — using the same SHA-256 hash chain that satisfies HIPAA §164.312(b) and 21 CFR Part 11 §11.10(e).

April 29, 202610 min read

The Legacy-to-FHIR Migration Playbook — Practical Patterns for Healthcare Data Modernization

Migrating from HL7 v2, X12 EDI, and proprietary formats to FHIR R4 is the defining challenge of healthcare IT. Here's a practical, phase-by-phase playbook for getting it right.

April 27, 20269 min read

FedRAMP & Public Sector Data Governance — What Federal Mandates Mean for Metadata Platforms

Federal agencies are modernizing data infrastructure under strict compliance frameworks. Here's how FedRAMP, FISMA, and public sector data mandates reshape the requirements for metadata management.

April 25, 20269 min read

Data Mesh Needs Governed Mappings — The Missing Coordination Layer

Data mesh promises domain ownership and decentralized data. But without governed mappings between domains, decentralization becomes fragmentation. Here's the coordination layer most implementations are missing.

April 23, 20269 min read

FAIR Data Compliance Scoring — Measuring What Makes Data Truly Reusable

The FAIR principles (Findable, Accessible, Interoperable, Reusable) are the gold standard for data quality — but most organizations can't measure compliance. Here's how automated metadata analysis changes that.

April 22, 20269 min read

Inside the Semantic Match Engine

A technical walkthrough of how Magellan turns two schema field paths into a ranked, calibrated proposal — embedding, calibrated cosine ranking, and confidence calibration, end to end.

April 21, 20269 min read

The ROI of Governed Metadata — Quantifying the Business Case for Metadata as Code™

Metadata governance isn't just good engineering — it's a measurable business advantage. Here's how to quantify the time savings, error reduction, and compliance cost avoidance.

April 21, 20268 min read

Data Observability at the Mapping Layer — Catching Quality Issues Before They Cascade

Most data observability tools monitor tables and dashboards. By the time they catch a problem, the damage is done. Here's why observability belongs at the mapping layer — where data is actually transformed.

April 21, 20266 min read

Predictive Mapping — Why We Don't Say "AI‑Powered"

Predictive mapping is a deterministic pipeline of proposal, named-SME approval, and signed commit. Here is how Magellan's Semantic Match Engine differs from autonomous AI mapping — and why regulators accept it.

April 20, 20268 min read

Why Three Roles Aren't Enough: RBAC for Regulated Audit Ledgers

The standard SaaS reader / contributor / administrator model violates separation of duties for regulated audit ledgers. Magellan uses a seven-role model that maps cleanly to HIPAA §164.308(a)(2) & §164.312(b), 21 CFR Part 11 §11.10/§11.200, SOX §404 SoD, EU GMP Annex 11, GDPR Art. 32, FedRAMP AC-5/AC-6, and ISO 27001 A.5.3 / A.5.18.

April 20, 20265 min read

Magellan's Audit Architecture, in One Page

A short tour of the four moving parts that make Magellan's audit ledger defensible across HIPAA, 21 CFR Part 11, SOX §404, EU GMP Annex 11, GDPR Art. 30, FedRAMP (NIST SP 800-53 AU family), and ISO 27001 A.12.4: ULID-keyed entries, SHA-256 hash chain, role-at-time-of-action, and two-component electronic signatures.

April 20, 20265 min read

What an Auditor Actually Sees in Magellan — Across HIPAA, SOX, GxP, GDPR, FedRAMP, and ISO 27001

A short walkthrough of the questions an external assessor asks, the queries Magellan runs, and the evidence pack exported — from the assessor's seat. The same five questions come up in HIPAA OCR audits, SOX walkthroughs, FDA Part 11 inspections, GDPR supervisory-authority requests, FedRAMP 3PAO assessments, and ISO 27001 surveillance audits.

April 20, 20265 min read

Audit Evidence Export and Periodic Attestation in Magellan

How Magellan's Compliance Officer exports a verifiable evidence pack and applies the periodic attestation electronic signature that satisfies 21 CFR Part 11 §11.10(e)/§11.10(j), SOX §404 management assertion, EU GMP Annex 11 §9, GDPR Art. 30, FedRAMP AU-6, and ISO 27001 A.5.35.

April 20, 20269 min read

Why the Magellan Audit Ledger Lives in the Same Container as magellan-data

Why we co-located the audit ledger writer inside magellan-data instead of building a separate audit microservice — the four structural splits that make it defensible under HIPAA, SOX, GxP, GDPR, FedRAMP, and ISO 27001, and the three conditions that would justify a split later.

April 19, 20268 min read

Data Governance Isn't Optional — How Metadata Management Keeps You Audit‑Ready

HIPAA, GDPR, CCPA, SOX — regulations demand demonstrable data lineage and provenance. Here's how metadata governance turns compliance from a fire drill into a byproduct of good engineering.

April 19, 20268 min read

Data Contracts & Schema Evolution — Taming Change Across the Data Pipeline

Schemas change. Pipelines break. Data contracts are the missing agreement layer between producers and consumers — here's how to implement them without slowing your team down.

April 19, 20266 min read

Why Magellan Uses ULIDs Instead of UUIDs for the Audit Ledger

ULIDs give Magellan's audit ledger something UUIDs cannot: a chronologically self-ordering, lexicographically sortable, distributed-safe primary key that maps cleanly to HIPAA §164.312(b) and 21 CFR Part 11 §11.10(e) audit-trail requirements.

April 19, 20264 min read

Hash-Chain Integrity for the Magellan Audit Ledger

A short technical walkthrough of how Magellan's SHA-256 hash chain detects tampering in the audit ledger and how a verifier reconstructs the chain to satisfy 21 CFR Part 11 §11.10(e).

April 17, 20268 min read

AI in Data Integration — Beyond the Hype, Into the Pipeline

AI is transforming data integration, but not in the way most vendors claim. Here's where AI actually delivers value — and where it still needs human oversight.

April 17, 20269 min read

FHIR & Healthcare Interoperability — Why Data Mapping Is the Missing Piece

Healthcare systems generate more data than ever, but interoperability remains elusive. Here's how metadata-driven mapping bridges the gap between FHIR mandates and real-world integration.

April 16, 20268 min read

From UIMA and TEI to the Semantic Match Engine — 20 Years of NLP Heritage Behind Magellan

Magellan's Semantic Match Engine runs on Gemini 2.5 Flash, but the design choices behind it — typed annotations, decomposable confidence, structured metadata — trace back two decades to Apache UIMA, Stanford CoreNLP, and the Text Encoding Initiative.

April 15, 20268 min read

5 ETL Mapping Mistakes That Break Pipelines — And How to Fix Them

Data mapping is the most error-prone step in any ETL pipeline. Here are the five most common mistakes teams make — and the practices that prevent them.

April 13, 20267 min read

Beyond String Matching — How the Semantic Match API Understands Your Data

Traditional field matching relies on string similarity. The Semantic Match API goes further — using AI to understand context, domain knowledge, and hierarchical relationships between fields.

April 11, 20266 min read

Metadata as Code™ — Why Your Mappings Deserve the Same Rigor as Your Application Code

Data mappings are critical infrastructure, yet most organizations treat them as throwaway spreadsheets. Metadata as Code™ changes that by applying versioning, CI/CD, and lineage to every mapping.

April 10, 20267 min read

Knowledge Graphs for Data Lineage — Why Flat Mapping Tables Aren't Enough

How Magellan's persistent knowledge graph tracks data lineage across transformations, unlocking impact analysis, regulatory compliance, and institutional memory at scale.

April 9, 20265 min read

Introducing Magellan — A Metadata‑Driven Data Mapping Platform

Today we're launching Magellan, a platform that treats data mappings as first‑class assets with strong governance, automation, and a persistent knowledge graph.