Service
AI integration & AI search visibility
Add Claude, OpenAI or Gemini to your product or workflow, from assistants to automation, and make your business visible in AI search.
AI is most useful when it does one job well inside a workflow your team already has.
AI features I can build
- Assistants that answer from your own content: help centres, product catalogues, policies and documents
- Customer support helpers: draft replies and route tickets
- Content tools: pages, blog posts, product descriptions and translations, with a review step
- Document reading: pull data out of PDFs, invoices, forms and spreadsheets
- Smarter search: understands what people mean, not only the exact words
- Classification and tagging: leads, tickets, products or messages
- Summaries and reports: from long text, calls or data
- Image generation: covers and product mock-ups
- Workflow automation: AI steps inside your existing app, run on a schedule or a trigger
- MCP servers: customers connect your product to Claude or ChatGPT
Built to be trusted
Cost limits per feature, a log of every call, fallbacks between providers and a human review step where it matters.
AI search visibility (GEO)
People now ask ChatGPT, Perplexity and Gemini for recommendations. I set up the technical side so AI assistants can read and cite your site: structured data, llms.txt, clean Markdown versions of key pages and crawler access.
Related work
Phuket real-estate marketplace with map search, lead tracking and an AI landing-page builder.
- Next.js 14
- TypeScript
- PostgreSQL
- Prisma
Frequently asked questions
What can AI do in my business?
Answer questions from your own content, draft support replies and content, read documents, tag and route requests, summarise, search more intelligently and automate repetitive steps.
Which AI provider should I use?
It depends on the task, price and data rules. I usually build behind one interface so you can switch between Claude, OpenAI and Gemini without rewriting the product.
What is GEO?
Generative engine optimization: making your site easy for AI assistants to read, understand and cite, so they recommend you when people ask.
How do you control AI costs?
With per-feature budgets, caching, smaller models for simple steps, and a log of every call with its token use.