Shiva Challa — Senior MDM Architect & Data Integration Lead

Published — Amazon KDP

When Your Systems Can't Agree

An Executive Field Guide to Master Data Management Using the Trust-First Framework

The field guide I wish I'd had on my first MDM project: a practitioner's framework for building data trust through governance, alignment, and operating model — before the next acquisition, audit, or AI initiative forces the issue.

Drafted with AI assistance and grounded entirely in two decades of hands-on enterprise work. The framework, the war stories, and the judgment calls are mine; the tooling just helped me get them onto the page faster.

$9.99 Kindle · $12.99 paperback · Free on Kindle Unlimited

View on Amazon → Explore the book site →

ISBN 979-8-1821-1006-8 · Independently published

What I do

Three problems I'm trusted with.

Most of my work lives where business-critical systems have to agree on the same facts — people, clients, vendors — and usually don't.

Master Data Management

Golden records, match-and-merge, survivorship, and stewardship workflows across People, Client-Matter, and Vendor domains — with the governance to keep them clean after launch.

Integration platforms

Centralized data hubs that replace brittle point-to-point feeds: REST/SOAP APIs, message queues, and direct access — engineered to be self-healing, with retry, reconciliation, and error recovery built in.

Cloud modernization

Moving legacy estates to Azure and Microsoft Fabric — Terraform infrastructure-as-code, Azure DevOps CI/CD, and medallion-architecture lakehouses — without disrupting what's already running.

Experience

Two decades, four platforms that mattered.

Hogan Lovells

2016 — Present

Washington, DC · 7th-largest law firm worldwide

Senior Data Architect — Data Integration Hub & MDM

Technical lead and architect of the firm's Data Integration Hub, the central backbone serving 44 downstream systems at 99.9%+ uptime — with self-healing built in: failed record updates are tracked and automatically retried on the next run, no manual intervention. Exposed to consumers via direct database access, REST and SOAP APIs, and controlled Excel access.
Built the enterprise MDM solution across People, Client-Matter, and Vendor/Payee domains; led teams ranging from 4 to 14 — including cleansing and de-duplicating client data against Dun & Bradstreet ahead of the Elite 3E migration.
Playing a key role in the firm's merger with Cadwalader — the largest law-firm combination in history — by migrating Cadwalader's downstream integrations into DIH.
Built the user-provisioning system (Workday → AD → Elite 3E) that now runs firm-wide onboarding, eliminating manual work; key role across four prior firm mergers.
Drove modernization to Azure and Microsoft Fabric — Terraform, Azure DevOps, and a medallion-architecture lakehouse for spend analytics.

Merkle

2013 — 2016

Columbia, MD

Senior Data Engineer — Customer MDM & Big Data Platform

Part of the team that built and ran the enterprise customer reference base — the Connected Recognition / CDI platform — Master Data Management at petabyte scale, the golden source other marketing platforms attributed campaigns against.
Operated and tuned the underlying 1.2 PB Hadoop data lake (Cloudera/HortonWorks) for Fortune 500 marketing analytics.
Built a reusable automation framework that cut client onboarding from 40+ hours to under two; onboarded T-Mobile, PayPal, ESPN, Geico, Toyota, and more.
Company-wide One Team Award for the Cloudera-to-HortonWorks migration.

Lockheed Martin

2009 — 2013

Manassas, VA · Energy Services

Data Warehouse / ETL Architect

Technical lead for the EDW/BI platform serving a utility of 150,000+ customers (J.D. Power #1, 2012).
Architected a self-contained MDM system in Oracle and the integration layer across WMIS, GIS, finance, and outage-management systems.
Stood up SAS DataFlux for source-system profiling, cleansing, and validation — enterprise data-quality tooling alongside the custom MDM.

Sprint Nextel

2007 — 2009

Reston, VA

DBA & SME — DB Design, SSIS, SSRS

Ran a 0.75 TB decision-support warehouse for 20,000 users at a 98% uptime SLA — feeding executive decisions up to CEO level, including the analytics behind a rise in customer-satisfaction scores. Delivered through a clean SQL Server 2000→2005 migration and SOX rollout.
Built early data stewardship by automation: a metadata-management job that detected missing field values and emailed the responsible user a daily digest to fill the gaps — closing the loop without a stewardship tool.
Automated performance self-tuning by monitoring DMVs and triggering reindexing only when fragmentation warranted it; published the underlying scripts to the SQL Server community.
Cut an Oracle load from 13+ hours to under one with SSIS.

InPhonic / Simplexity

2005 — 2007

Reston, VA

Database Administrator — SQL, SSIS & SSRS

Managed a 60+ SQL Server estate (largest database 400 GB+, five active/passive clusters) for a high-transaction wireless e-commerce platform; guided 150+ developers on design, indexing, and query performance.
Automated the CEO's daily report end-to-end — previously hand-built every morning — and owned the broader executive KPI reporting in SSRS that drove call-center, sales, and purchasing decisions, with metrics feeding the firm's quarterly financial earnings reporting.
Automated SOX-compliance obfuscation of sensitive data across every database, and built an SMO-based job-scripting utility that later became my public JobScripter.

Earlier — NJ Office of the State Treasurer, Hamburger One, SRG, and other DBA & development roles, 2001–2005.

Master data, made trustworthy.

When Your Systems Can't Agree

One query, the whole story.

Three problems I'm trusted with.

Master Data Management

Integration platforms

Cloud modernization

Two decades, four platforms that mattered.

Certifications & training.

Things I build off the clock.

Personal site & writing

desklog