Kevin Welter.
ENDE

My AI employee

Conversations become reusable knowledge.

Gustav runs in my own business and is also available as a ready-made AI employee package in my AI Employee Framework community, where you hire him for your own business.

I hand Gustav the recording of a call. He pseudonymises it first, that is the fixed first step, not an option, then distils it into a pattern register: the questions my clients ask again and again, each paired with my clearest human-reviewed answer so far. From that register come a playbook, FAQ answers, raw material for content and a library of objections. What he does not deliver is a new opinion. He distils what I have already said myself.

Whisper runs locally. The raw transcript never leaves the machine.

Join the communityRead first: Learning Claude
Get Gustav in the community

Whisper local

raw transcript stays on the machine

10 to 20 calls

per reviewed batch

13 lessons

ready-made community package

Call analysis

How calls become a knowledge base

A worked scenario shows how much expertise sits unused inside conversations.

300 calls, 45 min. each

Worked example in the course

225 hours

Expertise distilled from it

Batches of 10 to 20 calls

Work step

Before further analysis, a fixed pseudonymisation pass runs: names become a role, companies become an industry, amounts become an order of magnitude, places become a region, and sensitive passages about health, family or private finances are removed, not replaced. While attribution or realistic reidentification remains possible, the extracts remain personal data. The reason is not excess caution: secretly recording words spoken in private is a criminal offence under German law (§ 201 StGB). The course supplies three practical consent paths with templates before a microphone ever runs.

The actual processing runs in batches of 10 to 20 calls, with a sample review after every batch, and time spent is measured on the first batch rather than estimated in advance. The 225-hour figure shows the potential scale. Any existing recordings without consent stay out entirely regardless, more on that in the next section.

Gustav for your business

Different conversations. A knowledge system built for you.

Gustav is not limited to coaching calls. You decide which recordings go in, which patterns he should find and what form the resulting knowledge should take.

How I use him

Distill recurring customer questions from calls

Gustav transcribes locally with Whisper, pseudonymises the content and builds a pattern register of questions, answers, objections and content material.

Adapted to you

Your conversation types and knowledge product

You define the sources, consent, pseudonymisation rules and output. The same workflow can produce a playbook, FAQ, objection library or internal knowledge base.

What you change

  • Type and source of recordings
  • Questions and patterns to find
  • Pseudonymisation and deletion rules
  • Playbook, FAQ or knowledge base

Other possible uses

  • Sales calls
  • Support calls
  • Expert interviews
  • Internal meetings

The process stays: Consent first, Whisper local, remove sensitive data, review in small batches and transfer only evidence-backed statements into the knowledge system.

A knowledge network that can grow with every reviewed conversation

How Gustav's pattern register becomes visible as a knowledge graph

Gustav does not simply collect transcripts. His register connects recurring questions, objections, your clearest reviewed answers and their evidence. A simplified knowledge graph makes that network visible and can feed FAQs, objection cards, a sales playbook and content.

The structure in the diagram is that of the pattern register, the data is demo data: real conversations stay protected and do not appear here. This is what the view can look like that grows out of Gustav's pattern register of your reviewed conversations, evidence by evidence.

Build the knowledge graph

The tools

Gustav has his own 13-lesson course in my community that shows exactly this chain:

01

Transcribe conversations locally

Whisper runs locally on the machine itself, only the pseudonymised extracts later go into cloud processing, the raw transcript never leaves the machine.

02

Pseudonymise consistently

A fixed pseudonymisation pass before further analysis: role instead of name, industry instead of company, order of magnitude instead of amount, region instead of place.

03

Analyze calls in batches

Processing in batches of 10 to 20 calls with a sample review after every batch instead of one giant run.

04

Keep deletion possible

A ledger that always shows which call belongs to which person, so a deletion request can actually be carried out.

What he can't do, where the line is

Without a clean legal basis, Gustav doesn't get to see a transcript at all. Consent first, then the recording, then the processing.

Pseudonymise first, analyse second, always. That is the fixed first analysis stage, not an option.

Existing recordings without consent stay out: either consent is obtained afterward, or the call stays private or gets deleted. There is no retroactive fix.

Your own answers, not AI wisdom: Gustav distills what I have said myself, and doesn't invent advice on top of it.

A deletion plan is mandatory, not optional: the ledger always shows which call belongs to which person.

Learn the process. Or load Gustav directly.

Standard

Standard: build it yourself

$5a month

Learn the structure, rules and approval gates, then adapt Gustav to your own business.

  • Understand the build step by step
  • Adapt rules and examples to your own business

Premium

Ready-made template

Premium: ready-made template

$149a month

Gustav is a ready-made package in the community.

  • Load Gustav as a proven employee folder
  • Adapt the template to your workflows and approvals
  • Community and course in one shared workspace
Join the community

As of October 29, 2026

Frequently asked questions about Gustav

Can AI write a meeting protocol?

Yes, a protocol is the simplest level: who said what gets recorded and summarized. Gustav does not stop there: he distils recurring questions across many calls and pairs each one with my clearest human-reviewed answer so far, instead of just retelling a single conversation.

Can conversations be transcribed with AI in a GDPR-compliant way?

Only with a clean legal basis in place beforehand: secretly recording words spoken in private is a criminal offence under German law (§ 201 StGB). Gustav's course supplies three practical consent paths with templates, then the recording runs locally via Whisper, and a fixed pseudonymisation pass comes before any further analysis.

How do you transcribe conversations automatically?

Through a local speech recognition tool like Whisper, which turns the recording into text without the raw data ever leaving the machine. Only after that does Gustav's actual work start: pseudonymise, then distil into a pattern register of recurring questions and answers.

Does Whisper really run locally, without the cloud?

Yes, in the course Whisper runs directly on the machine itself, the raw transcript never leaves it. Only once the fixed pseudonymisation pass has removed or generalised names, companies, amounts and places do the pseudonymised extracts go into cloud processing for the pattern register.

What sets AI meeting notes apart from a real analysis?

Meeting notes summarize a single conversation. Gustav's pattern register compares many conversations and shows which questions recur and which reviewed answer has produced the clearest conversation signals so far. That is a basis for a playbook, FAQ and content, not proof of effectiveness from a single call.

Want to know what's hiding in your own calls?

Gustav's template is in the community, with the pseudonymisation pass and consent templates ready to rebuild.

Join the community

Read on

Also:Build FAQs From Customer Conversations·Transcribing Calls: What German Law Says·Call Analysis with AI: The Pattern Register·Build an AI Knowledge Base From Calls·Build a Knowledge Graph From Calls·Automatic Meeting Minutes with AI·Have Minutes Written: ChatGPT, Teams·Transcribing Voice Recordings on Your Phone·Transcribing Calls in Microsoft Teams·Getting a Transcript Made: Your Options·Transcription Software Compared·AI Knowledge Management for Companies·Learn Claude: The Path in Five Stages·Hiring an AI Employee: The Process·Creating an AI Employee: How to Start·AI and Privacy: What the AI Gets to See·Claude Privacy and GDPR: What Applies·German Skool Communities: Overview·AI for Coaches: Practice Over Hype·Hire your first AI employee·Back to the org chart