AI-native engineering: I build AI products, and I build with AI.
AI-native means two things in my work. First, I build products where AI is the core: LLM features, document pipelines and automation in production. Second, I build with AI every day. Claude, MCP servers, custom skills and subagents are part of how I research, write, review and ship code. The AI makes the work faster. I stay responsible for what ships.
#who-it's-for
- ▸Founders who want an engineer already working the way modern teams will work
- ▸Small teams that want AI-assisted delivery without losing control of their codebase
- ▸Companies building AI products that need someone fluent in both the models and the tooling
#what-you-get
Claude Code as a daily engineering tool, with project context files so the agent follows your codebase's rules
MCP servers that connect the agent to the tools and services a build depends on
Reusable skills for repeatable work: reviews, QA, deployments and documentation
Subagents for parallel research, codebase exploration and review
Human review on every change: AI drafts, a senior engineer decides
I learned to program before AI coding assistants existed: documentation, Stack Overflow, and a lot of trial and error. That foundation is exactly why I can use AI to move faster without losing track of what the code actually does. Read my story
#how-it-works
Which tools touch your code and data, and what stays out of bounds.
Project instructions and conventions so the AI works inside your standards, not around them.
AI speeds up research, scaffolding and testing; every change still gets a senior review.
Where it helps, your team keeps the skills and context files for their own work.
#related-work
#faq
Does AI-assisted code mean lower quality?
Not if a senior engineer stays in the loop. I use AI for speed in research, scaffolding, refactors and reviews, and I read and own every change before it ships. The quality bar doesn't move. The time to reach it does.
Will you use AI tools on my codebase?
Only in the way we agree up front. If your code or data can't be shared with AI tools, I work without them on your project.
Can you set up an AI-native workflow for my team?
Yes. Project context files, skills and MCP connections are reusable, so your team can keep using what we set up during the build.
Planning a build like this?
A 30-minute call is enough to tell whether I'm the right fit.