/

AI Center Of Excellence Lead for Enterprise Distributor

How MSH Tech placed the AI Center of Excellence Lead who turned a $20B distributor’s AI ambitions into governed, adopted, revenue-relevant systems — in a regulated environment where "move fast and break things" is not an option.

5 AI systems

In production across three operating companies within 14 months

7 mo → 10 wks

Approved use case to production

68%

Weekly active Copilot adoption (from 31%)

0

AI-related findings in the annual exam cycle

The Challenge

The client is not a company with an AI problem. It is a company with four very different operating businesses — vehicle distribution, F&I product sales and administration, auto finance, and franchise services — each with its own data estate, its own regulator posture, and its own definition of "ready."

By late 2023 the enterprise had done what most large organizations did: stood up an innovation council, run a wave of generative AI proofs of concept, and generated a backlog of more than sixty proposed use cases. Vendors were in the building. Associates were pasting customer data into consumer chatbots. Legal had begun asking questions nobody had a documented answer to.

What the enterprise did not have was a single accountable owner who could sit with a business unit president and a data engineering lead in the same hour, tell them the same story, and be believed by both. The gap was not technical talent. The enterprise had strong data engineers and a credible cloud platform. The gap was a translator with delivery authority, someone who could kill a bad idea in front of the executive who proposed it, and ship a good one through a model risk review without losing a quarter to it.

The Solution

The client's first-pass job description asked for "AI/ML expertise and executive presence." MSH pushed back and rewrote the screen around evidence of production delivery inside a regulated environment, which got the search unstuck.

What MSH Tech screened for, and how. Production evidence, not pilot evidence. Candidates had to name a live system, its users, its failure mode, and who got paged when it broke. Portfolio discipline. Candidates who had never killed anything had never had a real budget. Governance fluency. NIST AI RMF, model risk management under SR 11-7 discipline, ECOA/Reg B adverse action explainability. Adoption mechanics. The BU-leader test. If the candidate could not make that person care, the candidate was out.

The placed candidate came out of a large regional bank's enterprise data organization, with four years spent moving machine learning out of the lab and into servicing and fraud operations. They then led a data science and applied AI function at a mid-market insurance carrier, shipping a document intelligence platform and, more instructively, shutting down two flagship AI initiatives that their own CEO had championed. That second detail was the reason MSH advanced them.

“The scoring rubric was not there to pick winners. It was there so that when I told a business unit president no, I was not the one saying no — the process was. That is what makes it survivable, and that is what makes it stick.”

— Lead, AI Center of Excellence

Be prepared before your hiring emergency hits.

Speak with our team now to get us on your approved vendor list. From there, we can move with urgency to keep you mission-ready for any operational needs.

Get a Consultation

More companies that have solved big hiring challenges

Attracting top clinical talent during a shortage

MSH helps a fast-growing palliative care company find the right talent at the right time.

See Results

Strategic Talent Planning for Ground-Up IT Workforce Development

MSH helps a leading private equity firm design an integrated talent acquisition process to ensure top quality candidates and a lean hiring process.

See Results

Large Scale Data Migration

MSH helps a leading logistics company with $4B+ in revenue based out of South Florida migrate data warehouse infrastructure to AWS cloud.

See Results