David Arnoux Builds a GTM Second Brain for…

David Arnoux recently shared something that caught my attention because it was equal parts bold and practical. He opened with:

Announcing something pretty special today.

and also putting myself out of a job...

That line is a dare to every GTM leader: what if the best operators could package decades of judgment into a system that helps teams execute faster, with fewer bottlenecks and less dependence on a single expert?

David explained he has spent 22 years in growth, built a company to over EUR 20M in revenue, trained 500+ teams on GTM strategy, and consulted from seed to Series C across Europe. Then he said the quiet part out loud: in the last months, he realized he could fit all of it into one system. And he did.

The real promise: AI that behaves like an operator, not a toy

What David is describing is not another prompt library. It is closer to an AI-enabled operating system for go-to-market work: a structured set of playbooks, workflows, and agents that connects to your actual tools and then works with your actual data.

That distinction matters.

Most teams trying to adopt AI hit the same wall:

They get isolated wins (a decent outline, a rewritten email) but not compounding leverage.
Outputs are generic because the model does not know their customers, positioning, or performance data.
The team does not have the technical confidence to operationalize anything beyond chat.

David’s angle is to remove those blockers with what he calls a "second brain for Claude Code" focused on GTM, designed so marketers, growth, and sales teams can be productive fast.

Why "no terminal" onboarding is a bigger deal than it sounds

David explicitly calls out the adoption gap: never opened a terminal? The system takes you step by step with visual onboarding, cheat sheets, and a WhatsApp community. He even claims you can run your first positioning workshop within an hour.

If you have ever tried to roll out an AI workflow internally, you know where projects die:

Setup friction (accounts, keys, environments, connectors)
Fear of "breaking something"
Unclear first use case
No shared language for what "good" looks like

A guided onboarding path is not just a nice-to-have. It is the difference between an AI experiment and an AI habit.

The architecture: playbooks + agents + your tools + your data

The part that will make most GTM leaders lean in is David’s description of how it connects to a real stack: Slack, Notion, Google Drive, CRM, analytics, Google Search Console, and SEO data tools like Ahrefs and DataForSEO.

In practice, that implies three layers:

A curated GTM knowledge base (his playbooks and workflows)
Execution agents (skills that do specific tasks)
Context and feedback loops (your docs, conversations, KPIs, and market signals)

When those layers are wired together, AI stops being "content" and starts being "throughput".

The skills that feel like magic (and why they work)

David lists specific commands that "blow people’s minds." I think they land because they map to painful, repeatable work where speed matters.

SEO and distribution skills

/keyword-map: clusters 200 keywords by intent
/content-gap: finds what competitors rank for that you do not
/pseo-builder: generates 50 programmatic SEO pages (comparison pages, glossary, use cases) with schema markup
/geo-audit: checks if major AI assistants mention your brand

These are high leverage because they compress weeks of analysis and production into hours. But the deeper point is that they are not random. They are connected to outcomes: demand capture, discoverability, and brand presence in AI-mediated search.

Competitive research that is actually usable

David also describes /research as launching 5 parallel agents on competitors and producing a full dossier in 20 minutes: pricing, positioning, hiring signals, tech stack, gaps.

That is the kind of output that becomes a weekly ritual:

Monday: refresh competitor moves
Tuesday: update battlecards
Wednesday: adjust messaging experiments

And because it is repeatable, it becomes a system, not a one-off deliverable.

Positioning that goes beyond "write me a tagline"

One line from David’s post is worth pausing on: he contrasts shallow prompting with a "proper positioning exercise with competitive context" and ties it to the attention economy.

This is where AI can help the most, if you use it correctly. Positioning is not a brainstorm. It is a decision.

A good workflow here forces inputs (ICP, alternatives, proof, constraints) and produces outputs you can test (claims, narratives, landing page angles, sales talk tracks). If you are working on scroll-stopping first lines and messaging angles, a small helper like a free hook generator can be useful, but the real win is having positioning upstream so hooks do not become empty hype.

Creative production that lowers the cost of "showing" not just "telling"

David’s /animated-gifs example is telling: create animated infographics for a business model, customer journey, or tech stack in minutes.

Even if you never ship a single GIF, the meta benefit is speed to clarity. When teams can visualize a concept quickly, alignment gets easier:

Sales understands the product story
Marketing understands what to highlight
Product sees where messaging diverges from reality

The compounding advantage: pipelines, not isolated prompts

The most important part of the post, in my view, is not any individual skill. It is the pipelines.

David shares one sales pipeline:

/icp -> /signal-scan -> /cold-dance -> /call-prep -> /proposal

That is a full path from "who should we target" to "here is the signed proposal." The specifics will vary by business, but the concept is powerful: chain skills so each step uses the output of the previous one, with your context baked in.

He also shares a content pipeline:

/content-calendar -> /blog -> /repurpose -> /linkedin -> /carousel

One input becomes eight outputs. That only works when your system knows:

what you sell
who you sell to
what you believe
what proof you can cite
what your distribution channels reward

Without those, repurposing becomes spam. With those, it becomes consistency.

Where GTM teams should be cautious (so the system helps, not harms)

When you connect AI to Slack, Notion, Drive, and CRMs, you also inherit real risk. If you are adopting a "second brain" approach, I would ask three practical questions before going all in:

Data boundaries: what is allowed to be accessed and what is not?
Output accountability: who approves customer-facing claims, pricing statements, or competitive assertions?
Measurement: what metrics define success (pipeline velocity, content throughput, win rate, CAC payback), and how often do you review?

The goal is not to slow down. It is to keep speed from turning into mistakes at scale.

The bigger takeaway from David’s launch

David Arnoux is essentially arguing that the future of GTM is "AI-native out of the box":

fast onboarding for non-technical operators
reusable playbooks and agents
deep tool integration
workflows that chain end to end
regular updates as the space moves

If that vision holds, the competitive edge will shift from "who has access to AI" to "who has the best GTM system wrapped around AI." And that is why his opening line about putting himself out of a job resonates: the job is not disappearing, it is being productized.

This blog post expands on a viral LinkedIn post by David Arnoux, Helping GTM Leaders & Founders Grow With GTM x AI | Fractional CxO | Building Linkedin Tools @ humanoidz.ai. View the original LinkedIn post →