Find out if your organization is actually ready for AI
A written go/no-go on the initiative you were about to fund, the blockers ranked in closing order, and the plan to clear them. About four weeks.
Most organizations start building AI on a foundation nobody has examined. The data turns out to be messier than the architecture diagram suggests, no one has agreed who approves a model going live, and nobody has named the team that will own the system in production. The project stalls around month six, and the post-mortem surfaces problems that were all visible at the start.
This assessment produces a decision instead: whether the initiative you are about to fund will survive contact with your current foundation, and what has to change if it will not. It runs about four weeks end to end.
What you leave with
A written go/no-go on your first initiative. You name the initiative at the start. You get a written verdict on whether it is viable on your current foundation, and where the answer is no, the specific conditions that would turn it into a yes.
A ranked blocker list. Every gap I found across the five dimensions, ordered by what has to close before the next thing can start. Ranking is the useful part, because most organizations already suspect their gaps and are stuck on which one to fund first.
A sequenced action plan. Each blocker with an owner against it and a rough sense of the effort involved, in an order that respects the dependencies between them.
A written finding per dimension. What I examined, what I found, and the evidence, so your team can check the reasoning rather than take the conclusion on trust.
A first-pass governance framework. Decision rights, the approval path for shipping something customer-facing, and how an incident escalates. I size it for where you are rather than for a bank.
The five dimensions I cover
Data infrastructure and quality. Where the data lives, who owns each pipeline, how fresh it is where a model would consume it, and whether you can trace lineage when you have to defend an output.
Tooling and vendor coverage. Your current AI and data licences, where tools overlap and one can go, where a real gap sits, and what your existing contracts commit you to on data retention and training use.
Security and compliance posture. How data reaches a model provider and under what terms, access controls on the systems an agent would touch, and your regulatory exposure by jurisdiction and sector.
Talent and organizational capability. Who owns AI today, whether that ownership survives the current owner leaving, and the distance between the team you have and the system you want to run.
Governance readiness. What policy exists in practice rather than on paper, who holds decision rights on model selection and release, and how an incident would escalate.
Scope limits
I do not score your maturity and I do not recommend vendors. I resell nothing, so where I find that a tool you already pay for covers the gap, that is what the document says.
Proof, not promises
All client stories →AI Employee Assistance Chatbot
Designed and built a chatbot for a 40,000-employee organization to address questions about policies, asset tracking, and other internal tasks. Integrated with an updated knowledge base with an admin panel and the ticketing interface to create and track support tickets.
Leading financial institution in India
AI Vision for Fraud Prevention
Built a semantic image search interface to detect and present potential fraudulent gold loan applications to auditors for real-time fraud prevention.
Leading gold loan provider
Auto-scaling ML Inference
Designed and deployed a dynamically auto-scaling application for low-cost inference of ML jobs on geospatial data using GCP Cloud Run.
Public markets investor in MENA
Questions
What do I walk away with? +
Five things. A written finding for each of the five dimensions, stating what I found and the evidence behind it. A ranked blocker list in the order the blockers have to close. A sequenced action plan with an owner against each item. A written go/no-go on the specific initiative you named at the start, including the conditions that would turn a no into a yes. And a first-pass governance framework covering decision rights, approval path, and escalation route.
How long does it take? +
About four weeks from first call to final document. Most of the work happens on my side, and the demands on your team are conversations and system access rather than a project workstream.
What are the five dimensions? +
Data infrastructure and quality, tooling and vendor coverage, security and compliance posture, talent and organizational capability, and governance readiness. Each one produces its own written finding, because in practice the constraint is usually concentrated in one or two of them rather than spread evenly.
What happens if the answer is no-go? +
You get that in writing, with the conditions that would change it. That outcome has happened, and it is far cheaper than discovering the same thing nine months into a build. The action plan then sequences around clearing those conditions, so a no-go is a route rather than a dead end.
We have already started building. Is this still worth doing? +
Usually more so. Teams six to twelve months in have already hit the blockers, so the output shifts from prevention to diagnosis: which of the five dimensions is the actual constraint, and what changes. If the honest finding is that the project is sound and the timeline was optimistic, that is what the document says.
Is this a maturity score? +
No. I produce no score and no five-by-five grid. Maturity models rank you against an average organization that does not share your data estate, your regulatory exposure, or your team. The output is your specific blockers in the order you should close them.
Who needs to be involved? +
Whoever owns your data pipelines, your tooling and vendor contracts, and your security posture. For the findings review, at least one person who can approve or cancel the initiative under review. If nobody in the room holds that authority, the go/no-go has no effect.
Let's talk
30 minutes, no slides. We'll work the specific decision you're facing.