// 01
Operating at production scale — solo
43.5K
requests / day across 15 live domains
160+
countries served
1,000+
domains rebuilt from web archives
60–90
sites provisioned / month, automated
I design and operate AI-automated, multi-locale web platforms — LLM content pipelines, automated provisioning, and analytics that actually track correctly across every locale.
I'm based on the Costa del Sol and I build production automation for the businesses serving the international money moving here — real estate, relocation, the agencies behind them. Same timezone, English-native B2B, and working software you can click, not slideware.
Selling Costa del Sol property to British, German and Scandinavian buyers means listings in every language, ranking in every market. I built an engine that does it in one pass — native copy for six markets with the SEO built in.
See it live · Polyhaus →Every enquiry from a relocating family is a lead you can't afford to drop. I build front desks that read, qualify and book inbound enquiries automatically — and chase the quiet ones until they convert.
You win the international client; I'm the reliable build partner behind the automation and web work — production-grade, owned end to end, handed over so it keeps running.
Let's talk →On the coast, on CET, working in English — happy to start with a small paid test task.
Production demos — real automation running live, not mockups. Each handles a real business problem end to end.
Turns one property into publish-ready listings for six markets — each written natively for its buyers (not translated), with SEO and structured data built in.
View live demo · Polyhaus →An AI front desk that turns an inbound enquiry into a qualified, booked appointment — triaging by urgency and handling it end to end, without a human in the loop.
View live demo · Reception →Reads, routes, and replies to incoming service requests, then chases the quiet ones until they convert — recovering leads that would otherwise go cold.
View live demo · Intake →I turn manual, repetitive work into production LLM pipelines that run unattended — with guardrails and real error handling, not toy demos.
Next.js (App Router) at scale on Vercel, DNS/CDN via Cloudflare. Automated provisioning and config-driven deploys across 100+ markets.
GTM/GA4 rollout across multi-locale networks: config-driven injection, GEO-correct container mapping, execution-level QA. hreflang / canonical / sitemap / robots done right. This site is instrumented exactly this way.
Rebuild and operate a large network of content sites — work that normally needs a small team — without it collapsing under manual overhead.
Automated provisioning (60–90 sites/mo), LLM content & processing pipelines on the Claude API, Next.js builds on Vercel, DNS/CDN on Cloudflare, and a config-driven GTM/GA4 + technical-SEO layer across every locale.
1,000+ domains rebuilt, 15 live properties serving ~43.5K requests/day to 160+ countries — maintained end-to-end by a single engineer.
I'm Oleksandr — a front-end developer (5+ years) turned automation and infrastructure engineer.
Over the past year I've built and run a network of 1,000+ content sites largely on my own: automating provisioning, content pipelines on the Claude API, and the analytics & SEO layer underneath. The interesting part isn't any single tool — it's that the whole thing runs as a system, maintained by one person at a scale that usually needs a team.
I work fully remote from Málaga, Spain, with comfortable overlap across UK/EU and US-morning hours.
Tell me what's eating your team's time. If it's repetitive and lives between an LLM and your stack, I can probably automate it.