Building the revenue architecture of forward-thinking companies (YC S23) - Growth 🚀

Experienced T-shaped Revenue Maker, I have been delighted to work from B2C to B2B companies, from inbound to outbound strategies, from growth marketing to sales enablement ...  always in a single goal of generating more revenue.

Intent Strategies featured in Usergems & Lonescale 🚀🚀

https://www.usergems.com/blog/champion-tracking-setup-guide
https://www.usergems.com/blog/how-to-implement-new-hire-sales-trigger
https://www.lonescale.com/articles/how-to-implement-job-offers-trigger

Passionate about learnings new things (aka lifelong student) 📚 - Working every day on elevating my IQ, EQ & TQ.

Skillset:

- GROWTH 🚀
- Revops
- Sales
- Analytics
- Conversion Rate Optimization

 Personality: Dynamic, "challenger", at ease in contacts, coach

Maxence de Villepion 🧱

Garbage In, Garbage Out.
Cleaning a CRM is a project. Keeping it clean is a system. Martin Thomas and Hugo Cornu show you how.

Maxence de Villepion  🧱

Co-founder at Cargo (YC S23) — Building the revenue architecture of forward-thinking companies 🚀

Co-Founder @Architect Ops | On transforme votre pipeline en revenus | Clay & Cargo Partner 🧱

Votre CRM est peut-être votre pire ennemi commercial. 
Jeudi 5 mars avec Hugo, on vous montre comment en faire votre meilleur atout grâce à Cargo.

La plupart des CRM qu'on voit chez nos clients sont dans le même état.
Plein de data. Mais inutilisable.

Doublons, infos périmées ou manquantes, 
Sales qui perdent 15-20 min par fiche à reconstituer manuellement ce que l'outil devrait leur donner en 1 clic,
Head of sales qui répartit les comptes à l'aveugle.

Résultat : plus personne ne fait confiance à l'outil. 

On va vous montrer comment on a transformé le CRM d'un client en une source de données fiables et dynamiques avec Cargo 🧱.

Au programme :

→ Cargo vs Clay : pourquoi on a choisi Cargo pour ce projet et dans quels cas c'est le bon choix

→ Enrichissement + dédoublonnage : les workflows exacts qu'on a déployés pour nettoyer et faire vivre la base

→ Cartographie du TAM : segmentation en 3 tiers + 5 signaux d'achat activés (champion tracking, nouveaux recrutements, croissance…) 

→ Sales-ready : boutons HubSpot en 1 clic + agent IA sur Slack pour zéro friction au quotidien

À la fin de ce webinar, vous repartez avec une vision claire de ce qu'un CRM bien architecturé peut faire pour votre équipe commerciale.

📅 Jeudi 5 mars de 11h30 à 12h30
Lien d'inscription en commentaire 👇🏼

Maintenir une data CRM propre et activable avec Cargo

Aurelien Aubert

Investor

ADHD + AI is an unfair advantage.

AI is the first technology that rewards ADHD traits.

Old model: focus on one task, finish it, move to the next.
Now: launch a fix agent, launch another for the docs, check the first output, correct it, rerun it.

You're not executing anymore.
You're orchestrating.

And if your brain naturally works that way, you're already ahead.

Aurelien Aubert 🧱

AiSDR just patented "generating emails with an LLM and responding to replies."

Good for them.

In 2023, that was genuinely exciting.
In 2026, every GTM tool does this out of the box.

If a patent is your competitive strategy, the product conversation is already lost.

Build a better product.
That beats any patent.

No hard feelings. Everyone plays their cards.

We just shipped the Cargo CLI.

2 weeks ago, we released our TypeScript SDK (@cargo-ai/api).
Today we're going one step further: All Cargo, from your terminal.

npm install -g @cargo-ai/cli

One command. Full access to your GTM infrastructure.

- Create a Tool. 
- Trigger a run. 
- Send a message to an agent. 
- Query your data models.

Why this matters:
→ Built for Claude Code, Cursor, and agentic workflows
→ Your AI coding assistant can now operate your entire GTM stack
→ No browser needed — script it, automate it, ship it

This is just the beginning. Exciting things coming.

https://lnkd.in/eGBVGsR7

I learned to code at 11.
I'm living the period I always dreamed of.

I can build faster than ever.
I feel like a kid again.

Went through the whole traditional path.
Engineering school, then dropped out.

Spent years building random apps.
Most of them ended up in the trash.
Did some freelancing to pay the bills.

Joined Spendesk early. 
I was an engineer back then.

Met Maxence on the growth team.
The GTM tools sucked.

Couldn't do what we needed.
So we rebuilt everything in-house.

That's when it clicked: we were power users who got frustrated enough to build our own solution.

That's usually the strongest signal there's a real product to make.

Went solo as a freelancer for a bit.
Tested things out.
Then Max and I decided to build it as a SaaS.

When you've already rebuilt the tools once because nothing else worked, you know exactly what's missing.

We just launched our npm package.

Official TypeScript client for the Cargo API: manage data models, run workflows, and build AI agents from Node.js. Use it in Cursor or Claude Code to trigger agents/tools, analyse data, or pull workflow and model analytics. One install, full types, less boilerplate.

https://lnkd.in/eyUWvNvX

The new test isn’t “does it run?”
It’s “does it survive production?”

You can vibe-code a GTM product in a weekend.

You can’t vibe-code:
- Connectors that don’t silently break
- Data that doesn’t duplicate or drift
- Retries that don’t corrupt state
- Permissions that don’t leak customer data

AI made code cheap
It also multiplied the surface area of failure.

Infrastructure becomes the product.

I talked with top SWEs in the Bay Area last month.

80% of their code is AI-generated.

They prompt, review, ship. That's the whole job now.

Then I talked with GTM engineers.

Still building Clay tables.

You have nothing of an engineer if that's your stack.

Vibe GTM is not a thing.

Stop trying to be a Clay expert.

Learn Claude Code. Write real code. Build real systems.

The ones who don't will be replaced by the ones who do.

AI increased the workload.

Nobody talks about that honestly.

14 automations. 8 integrations. 

Fridays spent debugging why the sequence broke.

Scale glue and you get more glue.

Engineers use AI to eliminate toil.

GTM teams use AI to stack more work on top of fragile systems.

Same hours. Different foundations.

Build vs buy? 

Building won.
But not the way people think.

AI made code cheap.
It didn’t make systems reliable.

So yes, teams will build their own GTM logic:
scoring, routing, enrichment, agents

Because that’s where the edge is.

But they won’t build:
- connectors that don’t break
- sync that doesn’t duplicate
- pipelines that retry safely
- schemas that don’t explode

That’s infrastructure.

GTM is going the way engineering already did:
- You own the code that differentiates you.
- You run it on infrastructure you don’t have to babysit.

If your GTM strategy is stacking SaaS tools, you don’t have leverage.
You have dependencies.

The next generation of GTM teams won’t have a better stack.
They’ll own their logic.

And run it on infrastructure built for GTM.

We’re all racing to build the next best thing in AI. 

But most of the world has never touched AI. 

Plumbers, dentists, local shops,… they’re running 2026 like it’s 2010. 

The tools exist. Nobody showed them. 

Master the gap. Sell it. Dominate. 

Big wins go to the ones who zoom out, not the first arrivals.

IBM dropped 10% the day AI proved it could refactor their code. 

If your product is just logic in a UI, and AI can rewrite it in a weekend, you don’t have a moat, you have a countdown. 

Claude refactoring COBOL isn’t hype. 

Forty years of “too risky to touch” just became accessible. 

Obscure stacks and legacy languages aren’t defensibility anymore. 

The companies holding value don’t own code. 

They own distribution, embedded workflows, and real operational data. 

The moat is the system around it.

Engineers built the thing replacing them. 

Agents run in parallel, ship faster than teams ever could. 

Integrations. Bugs. Data. Verification. 

But Taste decides what survives. 

No shitty products. No product without vision. 

AI gives speed. 

You give direction. 

That’s the combo that wins.

AI doesn't make all engineers better.
It makes the best ones far more productive. 

And the rest, replaceable.

10x engineer + AI = 100x engineer.
1x engineer + AI = 0.1x engineer.

The 100x runs agents in parallel while working on something else.
A full team without the team.

The 0.1x used to fix bugs, build integrations, ship simple features.
That's AI's job now.

The gap between engineers isn't closing.
It's compounding.

Clay's new pricing is a great deal. For Clay.

The problem is the limits.

40,000 data actions sounds big. It's not.
Here's what that actually means in the real world:

Enrich 10k leads with a 4-step waterfall. 40k actions gone.
Run a Clay agent on 5k records. You're already burning thousands more.
Sync data back to Salesforce, HubSpot, or your warehouse. More actions.

One decent GTM workflow and your entire monthly budget is gone.

Now the second illusion: "cheaper integrations."
Only a handful of actions actually got cheaper. The heavy usage stayed the same:
waterfall enrichment, people data enrichment, Clay agents, most serious GTM workflows.

So the headline pricing looks great.
But the moment you run real pipelines, the math explodes.

To be fair: splitting enrichment and orchestration is the right move. We did the same thing a year ago.
The problem is the implementation.

The people getting hit hardest are the exact ones who made Clay successful.
Agencies running enrichment at scale. Power users building complex tables.
Teams running multi-step pipelines. Companies pushing serious data volume.

The builders.

This pricing is optimized for enterprise contracts now. Which is fine.
But when a product starts optimizing for procurement, power users start doing the math.

If you're running high-volume enrichment, complex workflows, or a custom GTM stack, you'll feel this immediately.

To all my frenchies 🇫🇷

https://lnkd.in/eht5nBsa

Co-Founder @Youno | Faites de votre CRM un système de revenu prédictible | Clay & Cargo Partner 🧱

Julián MEDINA, Guillaume Josset and 6 other connections attending

taplio

The Battle

Maxence de Villepion 🧱

Aurelien Aubert

Followers

Posts

Engagements

Likes

Performance ScoreMaxence de Villepion 🧱's LinkedIn Performance

Performance ScoreAurelien Aubert's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Maxence de Villepion 🧱

Aurelien Aubert

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Maxence de Villepion 🧱's LinkedIn Performance

Performance ScoreAurelien Aubert's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreMaxence de Villepion 🧱's LinkedIn Performance