Services
← Back to the homepageAI Readiness Assessment
Find out where AI will actually pay off in your organization, where it’s risky, and what it will cost, before you commit budget to it.
Who it’s for
- Leaders who feel pressure to “do something with AI” and want an honest place to start
- Organizations where staff already use AI tools without clear rules
- Teams that tried an AI tool and didn’t get the results they expected
What you get
- An executive summary written for someone who reads nothing else
- A ranked map of every AI opportunity, scored on value, feasibility, and risk
- 3 to 5 recommended quick wins, with what each would take
- Findings on current AI use, including unofficial use
- Risk and governance gaps, with the guardrail for each
- Cost estimates for the top recommendations
- A 90-day roadmap with owners and outcomes
Want to see what that looks like? Here is a sample report prepared for a fictional organization.
How it works
- Week 1: listen. Interviews with 5 to 8 people across leadership and hands-on roles, a short anonymous staff survey, and screen-share walkthroughs of real work.
- Week 2: analyze. Every opportunity scored on value, feasibility, and risk, with rough costs for the strongest ones.
- Week 3: deliver. A written report and a readout with your leadership team to agree on next steps.
Good to know
- Conducted remotely. On-site sessions are available on request, with travel billed at cost.
- Vendor-neutral. I don’t sell software, and I’ll tell you when AI isn’t the answer.
- Your information stays confidential and is never entered into AI tools without your written permission.
AI Team Setup
A working AI team inside your own environment: models from the providers you choose, clear roles, a structured workflow, and a person approving every step that matters.
Who it’s for
- Teams using AI tools one conversation at a time and hitting the limits
- Leaders worried about cost, drift, or AI making changes nobody approved
- Organizations that want the benefit of several AI models without handing any of them control
- Regulated teams that need a record of who approved what, and what each model did
What’s included
- Model selection across providers, matched to each model’s strengths and cost, using your own accounts, keys, and agreements
- Roles for each model with real job descriptions and a clear definition of done, kept in a reusable library
- A structured workflow from planning through build, peer review, testing, acceptance, and release, with lessons learned carried into the next cycle
- Peer review between models, so no model checks its own work, and human approval before anything ships
- Spending limits per model and per project, set by a manager rather than the team
- A record of what each model did, what it cost, and who approved it
- A setup designed to work alongside the tools your team already uses, such as Jira or Azure DevOps, so nothing is tracked twice
- Written SOPs and training so your team runs it without me
How it works
- Discover. How your team works today, the tools in use, and where it breaks down.
- Design. Roles, workflow, review gates, and spend controls that fit your team.
- Set up. Built inside your own environment, kept local to you.
- Train and hand off. Your team runs a real piece of work through it with me alongside, then without me.
Built on the same approach as my AI Planning Room. Typical engagements run 4 to 8 weeks, depending on team size and tools.
A look ahead
Screen designs for the next version of the Planning Room, now in development: a workspace for the team, and a separate admin console where a manager controls models, budgets, and roles. Sample data is illustrative. Select a screen to view it full size.
AI Governance & Validation Review
An independent look at whether your AI-assisted processes will hold up to an audit, from someone who spent two decades inside regulated, audited systems.
Who it’s for
- Life sciences, healthcare, and financial organizations using or planning AI in controlled processes
- Quality and compliance teams asked to sign off on AI they didn’t choose
- Leaders preparing for questions from auditors, clients, or regulators
What it looks at
- Where AI touches regulated records, decisions, or customer outcomes
- Human review and sign-off: who checks what, and whether it’s documented
- Traceability from requirement to evidence
- Validation evidence for AI-assisted steps
- Context of use for each AI-assisted step, in line with FDA’s risk-based approach to AI credibility: the question the model informs, how much weight its output carries, and the consequence of an error
- Alignment with the NIST AI Risk Management Framework (Govern, Map, Measure, Manage), connected to the quality system and change control you already run
- Data handling, access, and what leaves your environment
- Vendor independence and conflicts of interest
This review identifies risks and gaps. It is not legal advice, and I recommend involving your counsel wherever law applies. The thinking behind it is the same as my Requirements Compliance Architect.