AI features inside your product
Extraction, summaries, assistants, agents that run a whole workflow, AI translation, content generation, research tools. I build AI features that live inside your product and work with your data, for companies with no developer of their own and for teams that want someone who has shipped this in production.
I'm all in on AI myself: I use it for my own work every day, and I have built products on Anthropic, OpenAI, Google and xAI models. I also know where the models fail, so I build in checks, limits and a way to review what they did.
I usually reply within one working day.
AI features I build
What they look like in a real product, not in a demo.
Extraction and classification
Pulling structured data out of emails, PDFs, invoices, listings or support tickets, with a schema the model has to follow and validation on the way in. The part that saves people the most time.
Assistants that know your data
A chat or search over your own documents, orders or knowledge base, with retrieval so answers come from your data and not from the model's imagination, and sources shown to the user.
Agents that run a workflow
Multi-step jobs: research something, draft it, check it, publish it. I have built a product that runs several agents per job across four providers, with the results tracked and billed per step.
Translation and content
AI translation with a workflow around it, generated summaries and descriptions, content that gets reviewed before it goes out. I replaced a failing translation service on a multilingual site with DeepL and Anthropic models.
Cost, limits and provider choice
Which model for which job, caching and batching so you don't pay twice, limits per user or per job, and a way to see what a feature costs to run. Provider-independent, so switching later is not a rewrite.
Validation and review
Models are wrong sometimes. Structured outputs, validation, fallbacks, and a review step where a mistake would cost you money or trust. The unglamorous part that makes a feature usable.
How I work
You don't need to have everything figured out before we talk. An idea, access to the repo, a screen recording or a list of problems is enough. I look at what is there, tell you what I would do and what it costs, and then I build it.
I use Claude Code, Codex and my own agent workflows for reading code, planning changes, writing tests, refactoring and debugging. It is not vibe coding. It is senior engineering with a much faster loop.
AI-first, in the product and in the work
I build AI features the same way I build everything else: with Claude Code, Codex and my own agent workflows doing the typing, and me deciding what should happen, checking the results and testing the edge cases. I have shipped products on Anthropic, OpenAI, Google and xAI models, so provider quirks are not new to me.
I stay in control
AI writes a lot of my code. I still make every technical decision, read every change before it ships, and test it. Fast, but meticulous: the speed is only worth something if what goes live is right.
The way you want to work
Some clients want to look at every change before it goes live, others want me to deploy. Some want a weekly call, most prefer written updates. I work well asynchronously, and I still ask the important questions when they come up.
I think like an owner
I still run my own online products. So when I build or fix yours, I think about traffic, conversion, revenue and support as much as about the code, and I tell you when something is not worth building.
Ways to work with me
Fixed price, a flat monthly rate, or hourly. It's always me doing the work, not an agency. If you're not sure, describe what you need and I'll suggest one.
Examples of my work
AI features I have shipped.
Workflows across four AI providers
A product in Laravel and Vue that runs several AI agents per job across Anthropic, OpenAI, Google and xAI, fed by large-scale web scraping, with team accounts, credits and subscription billing around it.
AI translation that replaced a failing service
I replaced a broken translation service on a multilingual property site with a workflow built on DeepL and Anthropic models, as part of a larger takeover and Laravel upgrade.
Reports written when the data changes
Users define workflows over market data, and the product writes reports with AI when something important happens, with validation, credits and Stripe billing kept inside the product.
What clients say
"Vincent does not just close tickets, he thinks through the product, the edge cases, and the customer impact."
"He doesn't need micromanagement - he's fully capable of making decisions and driving the project forward, which has freed me up to focus on other parts of the business."
"We're now rolling out the multilingual capabilities he built, which is exciting. Vincent is reliable, technically skilled, and a valuable partner for keeping our site running optimally and moving forward."
Questions
Which models do you use?
Whichever fits the job. I have shipped features on Anthropic, OpenAI, Google and xAI models, and I build so that switching providers later is not a rewrite. Cheaper models where they are good enough, the expensive ones where they earn it.
How do you keep the costs under control?
Cheaper models where they are good enough, batching, caching, limits per user or per job, and a clear view of what each feature costs to run. Credits or usage limits when the feature is customer-facing.
Is our data safe?
Your data stays inside your product and goes to the model provider only for the request itself. I explain what goes where, and we pick providers whose terms fit your requirements.
What happens when the AI is wrong?
It will be, sometimes. That is why I build validation, fallbacks, and a review step where it matters, instead of trusting the output blindly.
Can you add AI to an existing Laravel app?
Yes, that is the usual case. Most AI features are a well-placed addition to an app that already works, not a new product.
How do you charge?
Fixed price for a job with a clear outcome, a flat monthly rate for ongoing work, or hourly. Describe what you need and I'll suggest one.
Get a quote
What you want the AI to do, what data it would work with, and what your app is built on. A few sentences are enough. I'll reply with how I would build it and what it would cost.
If you would rather talk first, book a 30-minute call.