AI-assisted software development for your team
I set up AI coding workflows that take work from a Jira ticket to a reviewed pull request, with your tests and code review still in charge. I've done it for a national retailer's engineering teams.
- Pricing
- $300 an hour
- Where
- South Florida or remote
Where AI fits in your development process
Choose the tools
Claude Code, Cursor, Copilot or a mix. We test them on your code and check them against your security rules.
Set the rules
What code and data AI tools may see, where AI-written code needs extra review, and how secrets stay secret.
Build the workflows
Shared commands and skills that take a ticket to a plan, tests, code and a pull request, with a person approving the steps that matter.
Add AI to your pipeline
AI checks and fixes that run in GitHub Actions, with approval steps and a way to roll back.
Organize your docs
Put architecture notes and how-to guides in one place that both people and AI tools can read.
Train your team
Hands-on workshops on your own codebase, in person in South Florida or remote.
Your checks stay in charge
AI makes it easy to produce a lot of code quickly. The hard part is keeping that code correct, secure and easy to change. Your tests, code review and CI stay in charge, and AI works inside them.
From my recent work
- AI coding across a large engineering team
- At a national retailer, I introduced GitHub Copilot, then built Cursor commands and Claude skills that grew adoption across teams. Connecting Jira through MCP cut manual steps between a ticket and a pull request, and I added guardrails on what Claude is allowed to do.
- Software changes without a development team
- For small-business clients, I automated the whole development process. The client describes a change in plain words, and an AI principal engineer that knows the code and the product's direction writes it up in full. Separate agents build it, review it and fix what the review finds, then it ships to production with a one-click rollback. Gates based on each task's size and risk control what goes through.
Common questions
- Which AI coding tool should my team use?
- It depends on what your developers use today and what your security rules allow. Many teams use more than one: an assistant in the editor for everyday work and an agent such as Claude Code for bigger changes. I'll help you test them on your own code before you commit.
- Is AI-written code safe to ship?
- It can be, with the checks you already trust: tests, code review and CI. The real work is deciding where AI may act alone and where a person must approve.
- Will our code be used to train AI models?
- Business plans usually exclude your code from training, but terms and settings differ by tool and change often. I'll check each tool's current terms with you and set it up before your team starts.
- Where do we start?
- With a pilot: one team, a few workflows and clear rules. We widen it once it holds up.
- Do you run workshops in person?
- Yes, anywhere in South Florida. Remote workshops work well too. Working with me in South Florida
- Will you sign an NDA?
- Yes. I'm happy to sign a standard NDA before you share the details.
How we get started
Tell me about your team
Its size, the tools you use now, and what you want AI to help with.
I'll ask questions
About your codebase, security rules and how work moves from ticket to release. Answering them often sharpens the plan on its own.
Free call
If it looks like a good fit, we map out a pilot together on a free call. If your team doesn't need outside help, or a simpler setup will do, I'll tell you.
Agree on the pilot
We put the scope, goals, hours, first team and how we'll measure success in writing.
Start the pilot
Once the first invoice is paid, the pilot starts. We review the results together and roll it out to more teams once they're clear.
Tell me what you're working on.
A few sentences is enough. I'll reply within one business day with questions or a next step.