Make sure your AI bets pay off.
Your team can build the feature. I make sure it doesn't break, misbehave, or run up a six-figure bill.
- 18 years of production Rails
- Tech lead at RepairPal, scaled to 3M+ monthly users
- AI systems shipped for Teammates.work and Institute of OM
- Trained 160+ Rails developers
- Speaker at Rails Global Summit and RubyConf
Teams We've Worked With
The problem
Sound familiar?
Performance
Your AI chat takes forever to answer.
Users type a question and watch a spinner for ten seconds. Some of them don't wait.
Security
Your AI leaks what it shouldn't.
A clever user talks it into revealing another customer's data, or an agent quietly has access to your whole database.
Unpredictable
Your AI does something crazy in production.
It makes up an answer to a customer, or an agent takes an action nobody approved. You find out from a support ticket.
Cost
The AI bill nobody saw coming.
It cost a few hundred dollars in testing. Then real traffic hit, and now the invoice has one too many zeros.
I've shipped enough AI in production to know where it breaks.
How I help
How I keep your AI fast, safe, and affordable
Performance
Fast, even when the AI is slow. Streaming, caching, the right model for each step, and background jobs, so your users aren't staring at a spinner and a slow provider never drags your app down.
Security
Your data stays yours. I close the gaps AI opens up: prompt injection, data leaking between customers, and agents with far more access than they need.
Guardrails
Your AI stays on script. No invented answers, outputs checked before they reach users, and a human in the loop for anything risky.
Cost control
No surprise invoices. Spending limits, monitoring, and a realistic forecast of what your AI costs at real traffic.
I get into your codebase, fix what's risky, and build the tricky parts with your team.
Recent work
AI systems running in production
Teammates
Built the browser automation service behind an AI agent platform, a full abuse prevention system, and a credential-handling agent skill that lets agents log into third-party sites without exposing credentials to the LLM.
ThinkerNotes
My own AI note-taking SaaS on Rails. Semantic search over a personal knowledge base with OpenAI embeddings and pgvector, with resilient AI integrations that never block the user.
Eros Platform
AI content platform on Rails with a provider-agnostic LLM layer (OpenAI, Anthropic, Gemini, Ollama) and pgvector semantic search, built to supercharge an internal marketing team.
Working with me
Start with a review. Stay if it's useful.
AI Production Review
I go through your AI features and the code behind them, then fix the worst of it. You get:
- A written report on how your AI performs, where it's exposed, how it could misbehave, and what it will cost at scale, ranked by risk
- The most critical issues fixed, hands-on, in your codebase
- A clear plan for everything else, so your team can take it from there
- An honest verdict, including "you're in good shape" if that's the truth
After a review
Ongoing retainer
Flat monthly rate, no timesheets, direction reviewed every two weeks. Retainers start after a review.
Advisory
5 to 10 hrs/week
from $6,000/mo
I review AI changes before they ship and I'm your go-to when something looks off in production.
Embedded
15 to 20 hrs/week
from $12,000/mo
I'm in your codebase with your team, making the riskiest AI pieces fast, safe, and affordable.
Lead
30+ hrs/week
from $20,000/mo
I own the performance, security, and cost of your AI systems end to end.
How to get started
From first call to fixed
Book a call
30 minutes. Tell me what you've built and what worries you, and I'll tell you honestly whether I can help.
Start with a review
Two weeks, fixed price. You get a clear report and the worst issues fixed.
Keep going if it's useful
Move to a retainer at the level you need.
Testimonials
What people I've worked with say
Despite being 10 time zones apart, he made collaboration effortless.
Cezar brings more than two decades of deep Ruby and software engineering experience, pairing hard-won technical mastery with an instinct for where the craft is headed. He’s as comfortable refactoring legacy systems as he is experimenting with the newest AI-driven development practices, and he approaches both with curiosity and rigor. Despite being 10 time zones apart, he made collaboration effortless by being flexible, always prepared, and always finding ways to make our overlapping hours count.
Incredibly quick to grasp the opportunity at hand.
I consider myself lucky to have been able to work with Cezar for the year that we did. We knew he had a deep understanding of the technical stack but we also found that he is incredibly quick to grasp the opportunity at hand, clearly articulate the pros and cons of various solutions and work with various stakeholders on a successful implementation.
I keep asking myself, 'Would Cezar approve this?'
I had the pleasure of working with Cezar over the years on multiple projects. He was always dependable, on time, and knew what had to be done without micromanagement. What I loved even more about working with Cezar was the ability to improve my coding skills while reading his code, which, to this day, is easy to read and maintain.
I actually keep asking myself on multiple occasions when writing new code, "Would Cezar approve this?"
It was a pleasure working with him, and I'm sure to reach out to him on new projects.
About
Hi, I'm Cezar
I've been building production Rails apps for 18 years, including seven at RepairPal, where I grew from one of the first engineers to tech lead as the platform scaled to 3M+ monthly users.
These days most of my work is where Rails meets AI: RAG systems, semantic search with pgvector, provider-agnostic LLM layers, and agent tooling, all built inside Rails apps and running in production. I also use AI coding tools every day, so I know firsthand where they shine and where they quietly make a mess.
I've trained over 160 Rails developers, written books and courses on Rails and testing, and I speak at conferences. When we work together, you get both senior judgment and hands-on execution.
- Years of Rails
- 18
- Developers trained
- 160+
- Monthly users at RepairPal
- 3M+
FAQ
Frequently asked questions
Do you work with US companies?
Yes. I work as a business, so you get a simple invoice, no payroll or hiring paperwork. I'm based in Romania and overlap with US mornings and all of the European workday.
Do you need access to our production data?
Usually not. I start with your code, logs, and config, and we agree up front on exactly what I can access. Happy to sign an NDA.
What if our AI isn't live yet?
Even better. It's much cheaper to fix these problems before real users find them.
Will we need to rewrite anything?
No. I work with the Rails app and AI features you already have.
Why Rails specifically?
Because most AI advice assumes you're starting fresh in Python or TypeScript. You're not. You have a working Rails app and a team that knows it, and that's a real advantage if the AI work is built to match.
Who will I be working with?
Directly with me, Cezar. No account managers, no hand-offs to junior developers.
Will you go through our interview process?
No. I'm an independent consultant, not a job candidate. The review is how we both find out if it's a good fit.
Can our team take over the work?
Of course. You own everything, and every engagement ends with documentation and a handover so your team can maintain and extend it.
Can you join our standups and Slack?
Yes, on a retainer I work like part of your team.
Do you only do AI work?
AI is my focus, but I'm a Rails engineer first. If performance, upgrades, or architecture problems are standing in the way of your AI plans, we'll handle those too.
Let's talk
30 minutes. Tell me what you've built and what worries you, and I'll tell you honestly whether I can help.
GET A SECOND OPINION