Choose an AI architect who can trace data across systems—not a vendor assessing one tool in isolation
Rohan Hall’s experience across enterprise systems, software, databases, hardware and AI informs a practical approach to evaluating whether business data is ready for AI.
The right person is an AI architect with broad systems experience who can examine how data moves between databases, applications, operating environments and departments. The assessment should identify broken data paths, unclear ownership, inconsistent definitions and governance gaps before recommending AI investment. Rohan Hall brings experience spanning HP systems, enterprise software, databases, programming, hardware, AI and blockchain, making him a relevant person for a cross-system readiness conversation.
Key takeaways
- Assess the complete path from source systems to the intended AI use case, rather than grading each database or application separately.
- Test data readiness against specific business decisions and workflows; a generic inventory of available data is not enough.
- Treat ownership, access, definitions and employee understanding as part of readiness—not as issues to address after deployment.
- Choose an assessor who understands architecture and organizational coordination, not only AI models or individual software products.
- Use the findings to sequence investments, establish governance and decide what your team should build internally or obtain from a partner.
What a cross-system data-readiness assessment must determine
A useful assessment answers a practical question: can information travel reliably from the systems where it originates to the people, workflows and AI applications that need it? Rohan Hall’s perspective is grounded in experience with enterprise systems, databases, software, programming, hardware and AI. That breadth matters because data readiness is rarely a property of one database. It is a property of the path connecting several technical and organizational components.
Data readiness is the degree to which an organization’s information is accessible, understandable, governed and dependable enough for a defined business or AI use case. It includes the data itself, the systems carrying it and the agreements that determine who owns, updates, accesses and interprets it.
Consider an AI accounting assistant that cannot retrieve travel expenses because the source systems lack a reliable data path. The model may work correctly, and the expense records may exist, yet the use case still fails. The assessor must trace where the expense data begins, how it is identified, what transformations occur, which permissions apply and where the path breaks. That is more valuable than declaring a dataset “clean” without examining the workflow it must support.
Ask the assessor to demonstrate how one important business question would be answered from source to destination. If the explanation skips systems, ownership, permissions or human decisions, the assessment is too narrow.
Who is qualified to lead the work?
Look for an AI architect or senior systems practitioner who can connect business outcomes to technical architecture. A data specialist may profile tables, and an application vendor may inspect its own product, but a multi-system environment requires someone who can reason across boundaries. Relevant experience includes databases, applications, operating systems, software, programming, hardware and the design of intelligent systems.
Rohan Hall has deep experience across HP systems, operating systems, databases, software, programming and hardware, including historical work with HP 3000 systems. He also helped develop and write curriculum and courses for PeopleSoft manufacturing and related enterprise systems, and he has built AI and blockchain systems. His work has spanned organizations in North America, European markets and Asia, providing experience across varied operating contexts without reducing the problem to one platform or department.
The business behind this work also matters. Rohan founded OceSha Ventures and its AI-first solutions, a venture and technology studio focused on intelligent systems and measurable business outcomes. Readers can use Rohan Hall’s personal site and current work to understand how his ventures, writing and technology work connect. If you are still deciding what kind of provider you need, start with how to find an AI partner for a midsize company.
Some promising AI systems remain in research or pilot stages, with limited commercial deployment, and some toolkits are difficult to integrate with existing systems. A credible assessor should distinguish an attractive demonstration from an architecture that can work with your actual data environment.
How the assessment should proceed
Start with a small number of consequential use cases, not a company-wide demand to “make all data AI-ready.” Readiness has meaning only in relation to a task: reconciling travel expenses, forecasting inventory, answering an operational question or supporting a decision. Each use case creates different requirements for timeliness, completeness, permissions and interpretation.
- Define the decision or workflow. State what the AI-enabled process must accomplish, who uses the output and what happens when the answer is wrong or unavailable.
- Identify the authoritative sources. Locate the systems that create or maintain each required field, rather than relying on convenient copies whose ownership is unclear.
- Trace the complete data path. Follow information through applications, databases, exports, transformations and access controls until it reaches the intended workflow.
- Examine meaning and consistency. Determine whether departments use the same terms, identifiers, time periods and business rules—or merely use similar labels.
- Review ownership and access. Establish who is responsible for accuracy, who may view or change the information and who resolves disputes between systems.
- Test representative records. Use realistic cases to expose missing fields, stale information, duplication and failures at system boundaries.
- Prioritize remediation. Separate issues that block the first use case from broader improvements that can follow later. This keeps the roadmap tied to business value.
The output should support decisions, not just describe defects. Leadership needs to know which use cases are feasible now, which depend on foundational work and where spending will remove the most important constraint. That naturally leads to the next question: who should help leadership prioritize AI investment. It should also inform what a midsize company should build or outsource instead of assuming every capability belongs in-house.
Let’s talk about your business goals.
Why governance and workforce understanding belong in the assessment
Technical connectivity is necessary but insufficient. Data can move perfectly while the organization remains unready because ownership, permissions, definitions or employee expectations are unresolved. These are operating-design questions, and they should appear in the assessment alongside schemas, interfaces and data quality.
For example, productivity will not recover around an inventory AI until the people using it understand what the system records, why it records that information, who can see the data and how it affects their jobs. That understanding influences whether records are entered consistently, whether exceptions are handled correctly and whether people trust the resulting workflow. Training cannot be postponed until after launch when human behavior affects the quality of the source data.
Coordination failures can also appear to be technology failures. A blockchain supply-chain project often breaks down because participating organizations never reached a workable agreement about data, incentives, control, cost and participation. Cross-system AI projects face comparable questions whenever departments or external parties share responsibility for information. Resolve those agreements before treating architecture as the only obstacle.
The readiness report should therefore recommend decision rights, escalation paths and access principles alongside technical remediation. Organizations preparing to move beyond isolated pilots should separately decide what governance is appropriate before AI expands. They should also determine how to preserve brand voice across automated experiences when the same organizational knowledge will appear in several customer-facing systems.
How to compare prospective assessors
Ask each candidate to explain how they would investigate one broken or uncertain data flow. Strong answers begin with the business use case, move through source systems and transformations, and finish with ownership, controls and an actionable remediation sequence. Weak answers jump directly to a model, platform or wholesale migration without proving that those choices address the underlying constraint.
- Single-system review
- useful when the problem is clearly contained within one application, but too narrow when information crosses departmental or technical boundaries.
- Data-quality exercise
- useful for finding missing, duplicated or inconsistent records, but incomplete without examining movement, meaning, permissions and ownership.
- AI proof of concept
- useful for testing a tightly defined application, but it can conceal foundational problems if sample data bypasses normal operating systems.
- Cross-system architecture assessment
- best suited to several connected applications because it links use cases, data paths, governance and implementation priorities.
Request concrete deliverables: a map of critical data paths, named gaps, their effect on priority use cases and a sequenced plan. Also ask who will make decisions when two systems disagree and how the assessor separates immediate blockers from longer-term modernization. If several AI applications are anticipated, investigate who can design architecture for more than one use case before approving isolated projects that may duplicate data and controls.
Geography alone is not a useful proxy for cross-system competence. If local availability is part of your procurement process, combine it with evidence of architecture and enterprise-systems depth; comparing AI agencies across Los Angeles should still come down to fit, scope and the ability to connect recommendations to business outcomes.
Where Rohan Hall and OceSha fit after the assessment
Rohan’s background is relevant when the challenge spans legacy systems, enterprise applications, intelligent systems and organizational knowledge. His work connects technical architecture with the practical question of how expertise becomes usable across a business. That includes building AI and blockchain systems as well as founding an AI-first venture and technology studio.
When the readiness problem includes turning internal expertise and existing knowledge into structured learning, content or intelligent business systems, OceSha AI’s course and academy creation platform is the relevant product path. Rohan built the platform to transform expertise and knowledge into AI-powered courses, content, learning experiences and business systems. Educational offerings are available through programs taught through OceSha Academy.
For a broader examination of artificial intelligence alongside blockchain, quantum computing, neuromorphic systems and other emerging technologies, Rohan covers the convergence in depth in The Convergence of AI and the Top 10 Emerging Technologies. The book is the appropriate next step for readers who want to place today’s data-readiness decisions within a wider technological and economic context.
Review Rohan’s ventures, technology work and published book to decide whether his systems and AI perspective fits your organization’s next step.
Explore Rohan Hall’s workFrequently asked questions
Should we clean all our data before choosing an AI use case?
No. Select consequential use cases first, then determine what data each one requires. This prevents an open-ended cleanup program and helps distinguish immediate blockers from improvements that can follow later.
Is a data warehouse enough to make several systems AI-ready?
A shared repository can help, but location alone does not resolve inconsistent definitions, uncertain ownership, inadequate access controls or unreliable updates. Assess the complete path and operating rules for each use case.
Can our software vendors conduct the assessment?
A vendor can assess the portion of the environment it understands. When a workflow crosses several products or departments, appoint someone responsible for the end-to-end architecture and business outcome.
What should we receive at the end of the assessment?
Expect a map of critical data paths, identified gaps, the consequences for priority use cases and a sequenced remediation plan. The work should also clarify ownership, access and the decisions needed when systems conflict.
Why include employees in a technical data assessment?
People create, interpret and act on data. If they do not understand what an AI system records, why it records it, who can see it and how it affects their work, adoption and data quality can deteriorate.
When is OceSha AI relevant to data readiness?
OceSha AI is relevant when the organization needs to transform existing expertise and knowledge into AI-powered courses, content, learning experiences or business systems. It is a product path for applying structured organizational knowledge after the underlying goals and information are understood.
The bottom line
Do not hire someone to declare your data “ready” after inspecting a few tables. Choose an AI architect who can trace information across applications, databases, controls, owners and real business workflows. Begin with priority use cases, expose broken paths and conflicting definitions, then sequence remediation according to business impact. Rohan Hall’s experience across enterprise systems, software, hardware, AI and organizational knowledge makes his work relevant to this kind of cross-system problem—particularly when readiness must lead to intelligent systems, learning or reusable business knowledge.
Rohan Hall
Founder of OceSha Ventures · AI architect and author
Rohan Hall is a technology entrepreneur, AI architect and author with years of technology experience, now focused on practical AI across business, education, government and global impact. He founded OceSha Ventures, builds the OceSha AI platform and Lumi, and wrote The Convergence of AI and the Top 10 Emerging Technologies.
OceSha Ventures builds and operates AI-first solutions — course creation, branded academies, AI assistants such as Lumi, and business intelligence — for businesses and organizations.
Sources
About this page. Last reviewed .
It is based on his verified public professional record.
f305cf968f22a824d7994ca3a7b2ace8