AI Automation & Workflows
Processes that take care of themselves.
Workflows that run themselves: mail to CRM, quote to PDF to follow-up, your own chatbots on company knowledge. Built in n8n or Make, connected to your existing tools — documentation and team onboarding included.
[ In practice ]
What this looks like, concretely.
The practice that never closes
Practice test box (2:18, with sound): a small box I place in practices for a trial — online booking with PIN confirmation straight from the practice calendar, a digital waiting room without an app, and a phone assistant that answers calls, books appointments and sends reminders. No call gets lost after hours, no booking slips through the daily rush. Two weeks, no contract. Demo with a fictional practice and fictional patients.
My second brain
MHM Brain (0:42, with sound): one of my own automation tools — every note, client and project as a searchable 3D globe instead of scattered folders and chats. The Zentrale, my assistant John, answers questions straight from this knowledge and keeps tasks and appointments in view, by text or voice. Financial figures are redacted in the video.
My own Kanban, my own server
CoWorkboard (0:19, with sound): my own Kanban tool for client and project boards — checklists, due dates, members, comments and attachments, live on my own server instead of a monthly subscription. Exactly the feature set my team actually needs, without upsell tiers and without third-party servers holding my own data.
One trigger, one finished carousel
Example from n8n: weekly trigger → Claude researches & drafts the carousel → OpenArt generates the slides → text overlay & Cloudinary → approval via WhatsApp → publishing to Instagram. From topic idea to finished post, not a single manual click happens — except the approval itself.
The salesperson who never sleeps
Example from n8n: a WhatsApp sales agent answers customer questions straight from the product brochure (vector store, Claude, conversation memory) — and the same workflow publishes image or video with a generated caption automatically to Instagram, Facebook and LinkedIn. A question at midnight gets the same precise answer as one at noon.
Three routes, one bot
One workflow, three routes: a photo or video sent to the Telegram bot is cropped to 9:16, „/idee“ researches a topic from the DOCMA blog and news search and has an image generated from it, and every Monday morning a carousel is created automatically. Claude writes the captions, approval or scheduling happens through a form in the chat, publishing goes to Instagram and LinkedIn — one bot for three completely different content jobs.
[ What I build for you ]
Workflows, not duct-tape solutions.
Workflow automation
Built in n8n, Make or Zapier — whatever fits your stack. With error handling and monitoring, not just the happy path.
API & system integration
CRM, mail tools, accounting, website — cleanly connected via official APIs. No screen scraping, no shaky workarounds.
RAG chatbots & company knowledge
A chatbot that knows your real documents, your wiki, your support history — instead of generic answers out of nowhere.
Document & content pipelines
Quotes, reports, social content — generated and delivered automatically, with an approval step where it makes sense.
[ How I bill ]
Day rate, transparent.
Same day rate as vibecoding and AI content — automation is implementation, not a consulting premium. You pay for craft, not an agency markup.
[ Packages ]
Four tiers, clear steps.
Quick-Automation
1 day 1.200 €net · one-off- A single workflow
- E.g. mail → CRM or quote → PDF
- Setup + testing + handover
- Short docs
- 5 days delivery
Workflow package
3 days 3.600 €net · one-off- 3–4 linked automations
- Complete flow, e.g. inquiry → quote → follow-up → filing
- Error handling & monitoring
- 2 revision rounds
- 10 days delivery
Automation-Sprint
5 days 6.000 €net · one-off- Process mapping included
- Whole department workflow (e.g. onboarding, content pipeline)
- Multiple API integrations
- Team training at handover
- 3 revision rounds
- 3 weeks delivery
RAG / company knowledge system
8 days 9.600 €net · one-off- Your own chatbot on your documents
- Connected to Slack, website or e-mail
- Grounding & quality assurance
- Access rights & privacy configured
- 4 weeks delivery
Enterprise
ongoing on requestquoted individually- Multiple systems & locations
- Ongoing development
- Retainer model
- Performance SLA
- Custom delivery
[ How it works ]
Five steps, clear handovers.
Briefing & scope
Which process, which tools are in use, which goal. Fixed-scope quote.
Concept
Workflow design, interfaces, error cases. Short approval before building.
Implementation
Built in n8n or Make, API connections, continuous tests against real data.
Testing & fine-tuning
Edge cases, error handling, monitoring alerts before it goes live.
Handover & training
Docs, team onboarding, 30 days support included. Optional maintenance retainer.
[ Tech-Stack ]
Proven tools, no vendor lock-in.
[ FAQ ]
Questions & answers
All answers on AI Automation & Workflows — pricing, packages, process, hand-over — and 143 more questions on the FAQ page.
Briefly describe what repeats — I will get back to you within 24 h with a concrete proposal and time frame.
Goes well with
[ Social Stream ]
@mikehillebrandmedia
Latest reels and posts. Drag to turn, click to open.
I measure how the site is used (Google Analytics) and whether visitors come from my Facebook and Instagram ads (Meta Pixel) — only with your consent. Without it nothing is measured and no cookie is set for it. Details in the privacy policy