What I build, and what you get when it is done.
Four ways to work together. Each one ends with something running in production and a team that can operate it.
AI Architecture
Design the system before anyone touches a model. I map the data, the workflows, the cost per operation, and the ways it fails.
- Use-case selection and feasibility review
- Data and context architecture
- Model routing and cost modeling
- Build-versus-buy recommendation
- Written architecture brief your engineers can build from
Agentic Build
I write the code. Production agents and automation pipelines, deployed and observable.
- Multi-step agent workflows with tool use
- Automation across your existing systems
- Retrieval and business-context architecture
- Evaluation harnesses and audit trails
- Serverless and edge deployment
Handoff and Capability Transfer
The system is yours. I document it, train your team on it, and leave.
- Runbooks and architecture documentation
- Working sessions with your engineers
- Monitoring and cost dashboards
- Hiring guidance for the roles you now need
Technical Due Diligence and Review
An outside read on an AI system, a vendor pitch, or an acquisition target.
- Architecture and code review
- Cost and scaling analysis
- Vendor and build-plan assessment
- Written findings with a plain recommendation
How the work is structured.
Pricing depends on scope. I quote a fixed number before starting so you are not managing an open meter.
Two to three weeks
A fixed-scope architecture review and written recommendation. The lowest-risk way to start, and it stands on its own if you build it elsewhere.
Six to sixteen weeks
Design through production deployment, ending in handoff. Scoped and priced up front against a defined outcome.
Ongoing, monthly
Standing technical leadership for a team without a senior AI architect. Architecture decisions, code review, hiring, and roadmap.
By the session
Due diligence, second opinions, and board or investor-facing technical review.
Before you write.
Do you actually write the code?
Yes. Every line on an engagement is mine unless we have agreed otherwise in writing. There is no junior bench and no offshore delivery team. If that changes on a large project, you meet whoever joins before they start.
What if we are not sure AI is the right answer?
Then the assessment is the engagement. Part of the job is telling you when a database query, a rules engine, or hiring one person solves the problem better than a model does. I have talked clients out of AI projects and will do it again.
Who owns the code and the intellectual property?
You do, on delivery. That is in the agreement before work starts.
How do you handle our data?
Least access necessary, written into the agreement. I work in your environment where possible rather than moving data into mine. For regulated data I have prior experience with FDA-cleared medical software and healthcare systems, and I will tell you plainly where a model provider's terms create a problem for your compliance posture.
What size company do you work with?
Small business through enterprise. What matters more than headcount is whether there is a decision-maker who can say yes and a real operational problem to solve.
How much does an AI consulting engagement cost?
Pricing is fixed and quoted before work starts, so there is no open meter. A fixed-scope assessment runs two to three weeks. A full build runs six to sixteen weeks. Fractional technical leadership is billed monthly, and advisory work is billed by the session. The first conversation is free.
Where are you located, and do you work remotely?
New Orleans, Louisiana. Engagements run remotely nationwide, with on-site work when the project requires it. There is a natural concentration of work across Louisiana and the Gulf Coast.
Are you available for full-time or interim roles?
Sometimes. Fractional and interim technical leadership are both on the table. Ask.