Open to relocation worldwide5+ years shipping production apps for EU & US teamsFull ownership: code → cloud → launchRemote, across every timezone

AI Automation

Research, content and workflows that run themselves — built with n8n and LLMs, owned by you.

Capabilities

The tools, wired together properly.

Automation Platforms

n8nZapierMakeWebhooksCron Jobs

AI & LLMs

OpenAI APIAnthropic (Claude) APIRAGPrompt EngineeringVector DatabasesLangChain

Content & Research

Web ScrapingMulti-Source ResearchContent GenerationSEO Automation

Integrations

SlackNotionGoogle SheetsCRMsSocial Media APIsSMTP / Email APIs

Chat & Support

Chatbot DesignIntent RoutingKnowledge Base RetrievalHuman Handoff Logic

Flows

A few of the automations I build.

Simplified pipelines — the real ones have more error handling and edge cases than fit in a diagram. Five of these, written up for MENA SaaS teams →

Project Deep Research

Give it a topic and get back a structured, source-backed report instead of a scattered pile of open tabs.

Marketing Campaign Research, Creation & Review

Research, drafts and a human review step built into one pipeline, so campaigns move fast without skipping the check that actually matters.

UGC Content Automation (TikTok / Instagram)

Turns a content idea into a scheduled, on-brand post — script, captions and hashtags included.

Skilled Chatbot

A chatbot that actually knows your product, not a generic assistant bolted onto a website.

Newsletter Without Mailchimp

A self-hosted newsletter pipeline running in n8n — no per-subscriber fees, no vendor lock-in.

Why

Automation that stays yours.

Built on tools you actually own.

n8n and self-hosted workflows instead of a stack of SaaS subscriptions that charge per contact and lock your data in.

A human checkpoint where it matters.

Content and campaigns get a review step before they go out — not a blind auto-publish button.

Grounded in your own data.

Chatbots and research pipelines pull from your actual docs and sources, not just a model's general knowledge.

One person, the whole pipeline.

The person who designs the flow also builds, connects and maintains it — no hand-off between "the AI person" and "the dev."

Process

From "wouldn't it be nice if this ran itself" to a working automation.

  1. Discovery & Use Case

    Understand what you're trying to automate, and whether AI actually helps here or a simpler workflow is the better answer.

  2. Flow Design

    Map out the pipeline — triggers, data sources, decision points, and where a human needs to stay in the loop.

  3. Tool Selection

    Pick the right mix of n8n, LLM APIs and integrations for the job, not just the trendiest stack.

  4. Build & Connect

    Wire up the actual automation — triggers, API calls, prompts, and the glue between them.

  5. Testing & Guardrails

    Run it against real inputs and edge cases before it touches live data or a live audience.

  6. Launch & Handoff

    Ship it with documentation so you can see how it works and adjust it without needing me for every tweak.

  7. Monitoring & Iteration

    Automations drift as sources and APIs change — I keep an eye on it and improve it over time.