Data & AI Decision Session with James Gray

You have a high-stakes data and AI call to make. Maybe a vendor shortlist to cut (data platform, AI tooling, or both). A build/buy decision. A roadmap to commit. A board narrative to defend in the next 30 days.

Your inputs are partial: vendor pitches that all sound the same, internal opinions split three ways, scattered Substack reads, and a Slack channel of takes from people who haven't sat in your seat. None of it adds up to a defensible call.

The most senior person in your org on data and AI is probably you. That's a specific kind of exposure — sound too cautious and you look behind, sound too confident and you risk a costly miss. The landscape is moving faster than your reps in it, especially where data and AI intersect.

Senior leaders shouldn't have to fly blind on the most consequential data and AI shift of the decade. A one-hour conversation with the right operator should be a normal part of your decision toolkit — not a 6-month consulting engagement, not a $5K advisory retainer, not a Slack DM to a friend.

Most AI advisors talk about AI. I talk about data and AI. The hardest calls sit at the seam between the two.

Sometimes the question is on the AI side — model selection, agent design, evaluation, vendor lock-in. Sometimes it's on the data side — governance, joins, freshness, ownership. Often it's both, and which one you're really solving for changes the answer. The build/buy call isn't a model question or a data question. It's a question about which parts of the stack you're willing to own.

Across 20+ years deep in data and AI — a decade at Microsoft scaling analytics platforms, 30+ enterprise consulting rooms, 6,000+ executives at UC Berkeley Haas Exec Ed — I've sat across from hundreds of leaders at decision moments like yours. The questions vary; the shape of the moment doesn't. It's rarely about the tech. It's about whether your data foundation and operating model are ready, what the next 90 days actually need to ship, and which calls are reversible vs. one-way doors.

My lens: AI amplifies who you already are — your strategic instincts, your team's operating model, your judgment under uncertainty. But AI is only as good as the data feeding it. The hour isn't about teaching you AI. It's about helping you make the human call about how data and AI fit your context — together, not separately.

Credentials:

• 20+ years deep in data and AI — operator, advisor, and educator

• A decade at Microsoft as a data executive, building and scaling analytics platforms

• 6,000+ executives taught at UC Berkeley Haas Exec Ed across Data Strategy and AI programs

• Trusted by leaders at Microsoft, Google, Amazon, Capital One, and UC Berkeley

• Master of Information and Data Science (UC Berkeley) + Berkeley Haas MBA + Brown certified leadership coach

1. Book and brief. Pick a time on Calendly and share the question, context, and any artifacts (vendor list, roadmap doc, board deck, architecture diagram, eval results — whatever's relevant to your call) beforehand. I review before we meet so the hour starts at depth, not setup.

2. Meet for 60 minutes. A focused, peer-to-peer working session. We pressure-test your thinking — including the data foundation underneath your AI plans — surface the calls that matter, and rule out the ones that don't.

3. Walk away with clarity. Leave the call with an expert-validated path forward and concrete next steps you can act on this week.

Investment: $300 for the hour. Payment and scheduling happen in one step on Calendly.

You walk back with a clear point of view. Sharpened by an outside operator who's seen this play out across hundreds of decisions. The vendor shortlist gets cut in half. The data and AI roadmap defends itself in the next exec review. The build/buy call is one you can stand behind in writing.

More than the artifact: you stop second-guessing where the real bottleneck is — sometimes the AI layer, sometimes the data layer, often both. You leave with a frame — what's reversible, what's a one-way door, what the next 30 days actually need to ship across your data stack and your AI plans — and bring that frame back into every decision your team makes for the rest of the quarter.

The hour pays for itself the next time someone in your org asks "should we buy or build it?"

The next 30 days look like the last 30. No outside stress-test. No senior operator in the room. The vendor shortlist still has six names. The data and AI roadmap arrives at the next exec review with the same caveats. The build/buy call gets deferred one more cycle, and the initiative slips another quarter — not because the call is hard, but because no one senior enough on both data and AI is there to make it.

The cost isn't the $300 you didn't spend on the hour. It's the budget you commit to the wrong vendor (data platform or AI tool), the headcount you scope against the wrong build, or the political capital you spend defending a decision that wasn't pressure-tested.

Book your session and walk away with a clear path forward.