Machinery thatkeeps yourbusinessturning.
After 15 years running data platforms in semiconductor manufacturing, healthcare, hospitality, games, SaaS, compliance, and utilities, I now bring that work to you. Architecture, governance, cost, and the path from model to production, built to hold up when the business leans on it.
Seven industries. One discipline.
Every one of these had different regulators, different data volumes, and different ideas about what "done" means. What carried across was the platform work: governed storage, reliable pipelines, a semantic layer people trust, and a clean path from a model to production. I have built that machinery at fab scale, under HIPAA, on a casino floor, and inside live games.
- Built and scaled data architecture, governance, quality, and MLOps teams at 2K Games, serving multiple studios on AWS, Snowflake, and Databricks
- Data platforms: lakehouse and warehouse design, streaming, open table formats, and platform-neutral target states that avoid single-vendor lock-in
- Governance for regulated data: HIPAA, PCI DSS, SOX, GDPR, CCPA, and COPPA, with operating models that have real owners rather than a policy deck
- Cost and FinOps: warehouse and cluster economics, workload policies, unit economics leadership can read, and vendor escalation when the bill is the vendor's fault
- MLOps and applied AI: feature pipelines, model registries, CI gates, and agentic analytics grounded in governed metrics, separating the ROI from the hype
- 7industriessemiconductor, healthcare, hospitality, games, SaaS, compliance, and utilities.
- 70,000+datasetscataloged across Snowflake and Databricks in one metadata rollout, ahead of schedule.
- 99.8%uptimeon production pipelines feeding high-volume ML training for a SaaS MLOps venture.
- $425Macquisitionsupported by the growth analytics foundation I built for a regulated DTC health business.
- 30%liftin data discovery, quality, and reliability from SLAs, monitoring, and CI/CD gate checks.
- 40kcallstranscribed and analyzed by an AI listening pipeline, extended to real-time signals.
15+ years at MGM Resorts, Pharmavite, Micron, and others. Worked with1
Work with me
Whether you need a fast diagnostic, a build with a defined finish line, or a senior data leader in the room before you are ready to hire one, I plug into your team and leave the machinery running.
Four ways to work together.
-
A number moved and nobody can say why, or you are about to sign a large vendor contract. I read your architecture, cost data, and governance as they actually run, and hand you a written findings memo with a fix-first order.
Warehouse and cluster spend review, catalog and lineage gap analysis, or a second opinion on an AI vendor shortlist.
-
You know what you need and want it built properly. I design it, build it with your engineers, document it, and leave it running in your accounts. Your team owns it on day one.
A metadata catalog rollout, FinOps dashboards and workload policies, a semantic layer with governed metrics, or a model-to-production pipeline with CI gates.
-
Your team has the people, but you want a senior data and AI platform leader to stress-test the plan. Roadmaps, vendor negotiations, executive readouts, and hiring plans.
Build-versus-buy calls on AI tooling, org design for your first governance and MLOps functions, or reviewing the data platform line in the budget.
-
You need a data platform leader before you are ready to hire one. I set the roadmap with your leadership team, then the hiring plan and the metrics you run on, and hand them to whoever comes next.
The platform roadmap for a new product, hiring your first data engineer, or getting AI spend under control.
My core capabilities
Across all engagements, I bring expertise in:
Platform and architecture
- Lakehouse and warehouse design
- Snowflake and Databricks estates
- Streaming and event pipelines
- Open table formats and query engines
- Semantic layer and metric definitions
- Platform-neutral target states
- Vendor evaluation and negotiation
- Cloud cost and FinOps
Governance and quality
- Governance operating models
- Data catalog and lineage rollouts
- Data quality SLAs and monitoring
- Regulated data: HIPAA, PCI DSS, SOX
- Privacy: GDPR, CCPA, COPPA
- Access control and audit readiness
- Compliance and safety reporting
AI, ML, and the team
- MLOps from roadmap to production
- Feature stores and model registries
- Agentic and conversational analytics
- AI-assisted engineering and code review
- AI spend and ROI narratives
- Data org design and hiring
- Executive readouts and board prep
Selected work
Four platforms I led, what was broken when I arrived, and what changed. Company names are withheld. Every figure is one I can walk you through.
- Games
70,000+Cataloging 70,000 datasets across two data platforms
A publisher with several studios and two data platforms had no single place to find a dataset or see where it came from. A structured proof of concept picked the catalog, and the rollout finished early.
Read the case study - Healthcare
$425MThe data foundation behind a $425M acquisition
A fast-growing subscription health business needed governed data under HIPAA, PCI DSS, and SOX. The foundation supported its growth analytics and a $425M acquisition.
Read the case study - Semiconductor
99.8%Keeping ML training pipelines at 99.8% uptime
A semiconductor manufacturer's SaaS MLOps venture needed pipelines that customers could count on. They ran at 99.8% uptime while volumes grew across fabs.
Read the case study - Hospitality
25%Governed reporting across a dozen systems for a resort operator
A resort operator had a dozen marketing and finance systems and no agreed version of the numbers. A governed lakehouse cut report cycles by 25%.
Read the case study
How I work
-
01
Intro call.
Thirty minutes, free. We skip the jargon and find the actual problem. We define the goal, the constraints, and whether I am the right person to solve it. If I am not, I will say so and point you to someone who is.
-
02
Scope on one page.
The problem, what done looks like, the timeline, and the price, on a single page. If anything changes, the page changes first. No surprises on the invoice.
-
03
The build, in the open.
You see the work as it happens, in your repos and your accounts, with a short written update every week. I advise and I execute. Nothing is revealed at the end.
-
04
Yours to keep.
Everything I build is documented, production-ready, and runs under your credentials. When I leave, nothing leaves with me. We decide the next step, or I finish the job.
Let's talk
Book a call
Thirty minutes, no strings. Bring the number that moved, the bill that grew, or the roadmap you are not sure about. You will leave with a first read and a straight answer on whether we are a fit.
Email me
Or email sid@millwrightdata.com directly.
Straight answers to the questions buyers ask
Each piece answers the question in its first two sentences, then goes as deep as the problem needs.
Why is our Snowflake bill so high?
Usually because warehouses run longer and larger than the work needs. How billing works, the six usual causes, and the order to fix them in.
Do we need a semantic layer before we add AI to analytics?
If you want AI to answer questions about your metrics, yes. What a semantic layer does, why AI makes it urgent, and how to start with the metrics in your board deck.
When should we hire a fractional head of data?
When data decisions are piling up faster than anyone senior can make them, but a full-time executive is not yet justified. The signs, the first 90 days, and how it ends.
Is our data ready for AI?
Ready means you can find it, trust it, trace it, and control who sees it. Four checks, a five-question self-test, and what to fix first.