First call
Menu

Lab

Tried, shown, explained.

On stage I show how automation and AI agents work. Here you find my talks, workflows from competitions and Zeitgeist, my experiment with a magazine run by AI agents.

Florian Fürst speaking at Berlin AI Day 2026 in front of a full room

Talks

Stages

Coming up

  • 11 November 2026, Mannheim: talk on Zeitgeist at the Claude Code Meetup hosted by Accenture
  • 25 and 26 November 2026: mentor at an AI bootcamp for a private equity firm, on behalf of Pantomeno
  • Berlin AI Day 2026: “From Spreadsheet to Solution”, the Goethe-Institut case, from an Excel spreadsheet to a portal
  • No-Code Week 2026: a sovereign RAG agent, entirely without US providers
  • Expert video for BIEG Hessen: “Artificial intelligence: building and using agents”, from simple automation to AI agent (in German). Video on YouTube
  • n8n Meetup Rhein-Main, Wiesbaden: several times since 2025, on changing topics
  • AI Week 2025: an AI agent in practice
Florian Fürst in front of a full room at Berlin AI Day 2026
Florian Fürst speaking at No-Code Week 2026
Florian Fürst presenting his school email workflow at the n8n Meetup Rhein-Main in Wiesbaden
Florian Fürst showing an n8n workflow at AI Week 2025

Looking for a speaker for your event? Write to me.

From the workshop

Workflows to look at.

I only show client projects with permission. I built these workflows for talks, competitions and challenges, so I can show them in full.

Refund Tracker

Second place at Flow4Gold at the n8n conference In The Loop 2026: three AI agents follow returns all the way to the refund. They read the drop-off receipt, check delivery with DHL, and look at the bank account and PayPal to see whether the money has arrived. If it has not, they suggest a friendly reminder to the retailer. It only goes out once I approve it.

  1. Capture the receipt

    1. Receipt uploadedGoogle DriveA photo or PDF of the drop-off receipt lands in a folder.Example: A Deutsche Post receipt: a DHL return to an online shop.
    2. Read the receiptPixtral · Mistral · AdacorThe image becomes text, and the text becomes carrier, tracking number, recipient and date.Example: Found: DHL, tracking number, shop, 9 September.

  2. Expect the refund

    1. Research the refund
      Research the refundGmail · Bank account · PayPalFinds the order confirmation and the expected amount. Delivery expected after one week, refund after two.Example: Order confirmation found: refund expected by 23 September.
    2. Add a rowData tableOne row per return, with amount and deadlines.Example: New row: open, delivery and refund dated.

  3. Check and act

    1. DHL agent
      DHL agentDHL APIChecks every day whether the return has been delivered, and records it.Example: Delivery confirmed.
    2. Refund agent
      Refund agentBank account · PayPal · GmailChecks bank account and PayPal. If the money is there, the case is closed. If the shop announces a date, it moves the deadline.Example: Deadline passed, no money yet.
    3. Approval
      ApprovalGmailThe agent proposes a friendly reminder. It only goes out once I approve.Example: Proposal by email, Florian approves.
    4. Reminder to the shopWeb search · GmailLooks up the shop’s service address and sends the reminder.Example: Reminder sent, the case stays on watch.

Simplified view of the n8n workflow. Refund Tracker, 2nd place Flow4Gold, n8n In The Loop 2026.

Sovereign RAG agent

For my class at No-Code Week Frankfurt 2026: an AI agent answers questions from your own documents. The files live in Nextcloud, the Supabase vector database is self-hosted, both on Hetzner servers. The models come from the IONOS AI Model Hub. Entirely without US providers.

Florian Fürst giving his class at No-Code Week Frankfurt 2026

All in Germany and the EU Servers at Hetzner, models via the IONOS AI Model Hub, Nextcloud and Supabase self-hosted. No US provider.

  1. Take in knowledge

    1. Schedulen8nStarts on its own at regular intervals.Example: Every morning at six.
    2. Fetch the filesNextcloud · HetznerDownloads the PDFs from a Nextcloud folder.Example: Three new manuals as PDF.
    3. Extract and split the textn8nTakes the text out of the PDF and splits it into sections.Example: 240 sections.
    4. Store in the vector databaseSupabase · IONOSEach section is stored as a vector, so it can be found by meaning.Example: Stored, findable by meaning.

  2. Answer questions

    1. Question in the chatChatSomeone asks a question.Example: “How often does the system need servicing?”
    2. AI agent
      AI agentIONOS · MemoryUnderstands the question, remembers the conversation and decides when to search.Example: The agent searches the manuals.
    3. Find matching sectionsSupabase · IONOSFinds the sections that match the question. The agent answers on that basis.Example: Two sections found, answer in the chat.

Simplified view of the n8n workflow. Sovereign RAG agent, No-Code Week Frankfurt 2026.

School and after-school emails

Built for my own family, a winner in n8n’s #FrAIday challenge in April 2026 and presented at the n8n Meetup Rhein-Main in May 2026: an AI agent reads the emails from school and after-school care, attachments included, puts dates in the calendar and tasks in the to-do list, and sends me a summary. The prize was an n8n hoodie, and I wore it at the meetup.

Florian Fürst in the n8n hoodie he won, at the n8n Meetup Rhein-Main

  1. Incoming

    1. Email from school or after-school careGmailCertain senders start the workflow.Example: The school writes about project week, with two attachments.

  2. Attachments

    1. Save attachmentsGoogle DriveEvery attachment is stored.Example: Two PDFs in the family folder.
    2. Analyse attachmentsGeminiReads the documents, so the agent knows what they say.Example: In the PDF: trip on Thursday, bring a lunch box.
    1. AI agent
      AI agentClaudeReads the email and attachments, checks the calendar and decides what to add.Example: One event, one task.

  4. Result

    1. EventsGoogle CalendarCreates events in the family calendar.Example: Thursday: trip.
    2. TasksGoogle TasksCreates tasks in the to-do list.Example: Wednesday evening: pack the lunch box.
    3. SummaryGmailSends a short summary by email.Example: A short email: what is coming up and what was added.

Simplified view of the n8n workflow. School and after-school emails, n8n #FrAIday Challenge 2026.

Support inbox: Inbox Inferno

For an n8n community challenge in March 2026: an agent sorts incoming requests by topic. Specialist agents for pricing, security and setup answer only from the documentation provided. Anything they cannot answer with confidence goes to a person. The models run in Adacor’s private cloud.

  1. Incoming

    1. Request arrivesWebhookAn email to support arrives as a record.Example: “How do I connect Salesforce?”

  2. Triage

    1. Triage agent
      Triage agentQwen · AdacorAssigns the request to a topic, following the escalation rules.Example: Topic: setup.

  3. Specialist agents

    1. Pricing
      PricingDocumentationAnswers only from the pricing and product documents.
    2. Security
      SecurityDocumentationAnswers only with approved responses on security and data protection.
    3. Setup
      SetupDocumentationAnswers from the product and integration documentation.Example: Draft reply from the integration guide.
    4. Escalation
      EscalationRulesWhatever cannot be answered with confidence goes to a person, with a polite handover.
    5. Spam and misdirected mailSorted out, no reply.

  4. Reply

    1. Reply goes backWebhook · GmailTopic and reply go back, as a record and by email.Example: Reply sent with the category “setup”.

Simplified view of the n8n workflow. Inbox Inferno, n8n community challenge, March 2026.

Zeitgeist · Experiment

A magazine run by AI agents

Zeitgeist is my agentic magazine about AI and what it means for people, and an experiment. AI agents monitor sources, research, check every number against the source and write analyses in German and English every day. I set the rules, the checks and the cost limits.

  • 5 desks and the n8n Lab
  • 3 to 6 articles a day, in German and English
  • More than 30 sources approved, from research to regulators
  • $100 a month cap for the language models

How it is built

Build

Claude Codedevelops, tests, deploys. Not part of daily operation.
Git repositoryWorkflows as code, prompts and rules, database migrations, theme.

Steering · AI agents

Editor-in-chiefdaily overview at 7:50 am, review on Mondays, approvals by email.
Watchdog · error alertat least 3 articles a day, every error by email.

Operation · n8n on Elestio (Hetzner, Falkenstein) · daily at 4:30 am

Sourcesmore than 30 approved: RSS, lists, Firecrawl.
ResearchKey statements with a verbatim quote.Sonnet
Desk leadsSyntheses first, then 5 desks select.Sonnet
Primary sourcefinds the original behind retellings.Sonnet
Counterpointlooks for a sourced opposing view.Sonnet
NewsroomGerman and English, with a self-check list.Opus
Standards checkNumbers, evidence, logic, DE = EN. Approves.Opus

↓ findings↑ check again

Reworkfixes findings, once.
approved
Publisherpublishes both languages.

↓

Expert requestTheir context appears afterwards.
Title card
Timeline
Operations database (Postgres)evidence files · tasks and check findings · approvals · own login per agent, only the check approvesBudget lock: at most $100 of model costs a month, every task reserves in advance.

Outside

WordPresszeitgeist.fuerst.ai, German and English, custom theme.
Anthropic APIClaude Opus 5.5 and Claude Sonnet 5.5.
Web and mailFirecrawl, Apify, zeitgeist@fuerst.ai.
LinkedIn agent3 carousels a week.prepared, waiting for access
Requests to sourcesopen questions by email.decided, not built
A human as publishersets the editorial line, sources and budget · decides on approvals and proposals by email · articles go out without prior review, experts add context afterwards
Agentic first: agents do the work, rules set the limits, a human is the publisher.

Articles are published without a human reviewing them first. Experts put selected articles in context afterwards. Decisions that matter are made by a human.

As an experiment, Zeitgeist is not finished: I am testing how far agents can do journalistic work, and I learn from every mistake.

Coming up

  • 11 November 2026, Mannheim: talk on Zeitgeist at the Claude Code Meetup hosted by Accenture

Read Zeitgeist

First call

What would your project be?

Tell me in 30 minutes where things get stuck.

Book a first call

Florian Fürst listening to a question from the audience at Berlin AI Day 2026
First call