AI builds fast.
It can't judge its own work.
Can you trust what you've built?
I'm a fractional CTO who uses AI coding tools daily — and brings 16 years of knowing what breaks. I build the workflows, guardrails, and conventions that let you and your AI make dependable progress together.
Not more code reviews. Not another meeting.
You're buying dependable progress: what to build, how to structure it, and where AI is likely to lie convincingly.
Architecture that won’t corner you
Clean boundaries, boring defaults, and reversible decisions — so you can keep moving fast without rebuilding every month.
Security, payments, and reliability
Auth that holds, Stripe that works, data that’s safe, deploys that don’t break. "Shipped" means "customers can pay you."
AI workflows that actually produce
Conventions, guardrails, and review patterns that make your AI a reliable teammate, not a liability.
Best fit: solo founders and small teams using AI to ship fast, who need a senior eye on what's going to production.
Build with AI the right way from day one
Free AI coding skills for Rails development — conventions, scaffolding, and setup patterns that save you from the mistakes AI makes when it doesn't know the framework. No signup required.
Rails is the best framework for AI-assisted development
Conventions beat configuration
90% of what you're telling your AI agent to build is already built into Rails. Less code means less surface area for AI to get wrong.
Ruby is token-friendly
Concise, readable, expressive. AI generates better Ruby than verbose languages — and you can actually read what it wrote.
Battle-tested at scale
Shopify, GitHub, Basecamp. Rails has 20 years of production answers to problems your AI will pretend are novel.
Want the full data? Read why Ruby is the best language for AI-assisted development.
Not on Rails yet? That's fine. Start with an audit — I've helped founders migrate from JS full-stacks when it made sense, and advised staying put when it didn't.
Codebase Review
AI-generated code looks right — that's the danger. A one-time deep audit to find out where you actually stand: architecture, security, performance, and the patterns that will bite you later.
You get a clear, prioritized report — not a vague list of suggestions. What to fix now, what can wait, and what's actually fine.
From €3,000
A full rewrite costs €50,000+ and 3–6 months. The audit tells you exactly where you stand.
What the audit covers
- Architecture & domain model quality
- Security vulnerabilities & auth hardening
- Performance bottlenecks & N+1 queries
- AI-generated anti-patterns & reinvented wheels
- Test coverage gaps
- Deployment & infrastructure risks
Avo case study: Lighthouse 81 → 98, response times nearly halved.
CTO Retainer
A senior technical partner in the room — not just once, but every week. I build the workflows, review the PRs, and set up the conventions that let you and your AI make real, dependable progress.
€5,000/mo
1 of 3 slots available“Julian reviews our code like a technical co-founder — containing risk, bringing architectural judgement, and giving me the confidence to ship fast.” — Naveed Akram, Indee.co
No long-term commitment. Cancel anytime.
What 20h/month looks like
- Async PR reviews through GitHub — ship without waiting
- Architecture guidance on every major decision
- AI workflow setup — conventions your agents follow
- Security & performance oversight, continuously
- Optional sync deep-work blocks for intensive shifts
- Direct access — no ticketing, no account managers
Not just a Rails expert. A trained eye for what AI can't see.
I'm a sound artist, software engineer, and generative AI researcher who describes himself as a “code sculptor.” I trained my perception on electroacoustic composition — separating what code does from how resilient it is. That artistic discipline is what I bring to every review.
StimulusReflex maintainer
Core maintainer of a foundational Rails real-time library — the code AI learned from.
16+ years in generative AI
Published book on algorithmic creativity (2008). Research associate at FH St. Pölten.
24+ published articles
AppSignal (author + technical editor), Cloud66, RoRvsWild.
Dozens of codebases reviewed
Rails audits across architecture, security, and performance.
Sound artist & composer
ZKM Karlsruhe, Ars Electronica, ORF Kunstradio. Trained perception applied to code.
“I recommend Julian to SaaS founders who’ve moved fast with AI tooling and are starting to wonder what they’ve missed. He’ll review your code and infrastructure to help you and your LLMs ensure it’s ready for business and growth.”
“It’s rare to find someone who can step into a fast-built, AI-assisted codebase and calmly turn it into something robust, secure and ready for real customers.”
“Julian is exemplary in going deep in the code to find the inefficiencies we overlooked. The fact that Avo is a library where the users have so much control makes this task even more challenging but we are grateful to have him review every aspect of it.”
24+ articles across leading Rails publications
Code Sculpture
A series on art, software craft, and what AI cannot understand about code. Seven essays on perception, entropy, and why the most important qualities of software are precisely the things machines can't replicate.
Free. Seven essays. No spam, no pitch.