Aurelien Aubert

building @cargo_hq

[Raytar @Raytar]

My friend makes $1.2 million a year as an Anthropic engineer.

I asked him how he learned prompting so well.

He sent me a video that was never supposed to get out. Their core team's prompting playbook.

You won’t find anything better about prompting than this video.

I watched it last night.

Halfway through, I realized I've been using Claude completely wrong for two years.

Watch it, then read the article below.

[Boris Cherny @bcherny]

As engineering, product, design, DS, etc. melt into a new kind of role, I was reflecting on what roles might look like in the future. For example, when I look at the Claude Code team I see what I think is five archetypes:

1. Prototyper: comes up with brand new ideas; churns out many ideas, most of which don't ship
2. Builder: quickly turns a prototype/idea into production-grade product/infra
3. Sweeper: cleans up the UI, simplifies the code and system, unships, optimizes performance
4. Grower: takes a product that has been built and iterates on it to improve Product-Market Fit
5. Maintainer: owns a mature system to make it secure, reliable, fast, and efficient as it scales

Many people span across 2 roles, and sometimes 3 roles. I also notice that these roles are not really tied to job function -- eg. across Anthropic, some designers match category 1, some 2, some 3; same for engineers, PM, DS.

A healthy team needs a mix of these, depending on the product:

- A product that is new and pre-PMF needs people that are strong at 1+2+3
- A product that is growing and has found PMF needs 2+3+4 and some 5
- A product that has strong PMF needs 3+4+5 and some 2

Maybe product roles of the future will look more like this, and less like the domain-specific roles of today?

[Aurelien Aubert @aureeaubert]

Poor France. Is it really how we’re supposed to celebrate?

[Will Marquis @will_marq]

Autonomous companies will become a reality before the end of 2026.

I want autonomous companies to be accessible for anyone with a computer, a dream, and Pancake.

Not just the top 0.1%.

Introducing Pancake: the OpenClaw cofounder that makes your company autonomous.

[Patrick Collison @patrickc]

Have wanted something like this for ages — it looks great.

[Aurelien Aubert @aureeaubert]

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.

[Boring_Business @BoringBiz_]

This 40 minute lecture from Peter Thiel at Stanford will teach you more about business competition than a 2 year MBA program

[Aurelien Aubert @aureeaubert]

It’s funny, the same people saying it’s not the end of the application layer are the ones who said AI wouldn’t replace engineering one year ago

[Aurelien Aubert @aureeaubert]

Salesforce won’t be killed by a better CRM.
The Salesforce disruptor will look more like AWS than a shiny UI.

Because the CRM era was about humans entering data.
The AI era is about systems producing truth.

Salesforce is an application:
- UI-first
- clicks as the runtime
- admins as the integration layer
- “custom objects” as a coping mechanism

AWS is infrastructure:
- primitives
- APIs
-guarantees
- boring reliability at massive scale

That’s what the next CRM has to be.
Not a place to log notes.

A revenue substrate:
- identity resolution + context graph
- event streams from every system
- deterministic pipelines (retry, idempotency, no duplicates)
- schemas + versioning (change without breaking everything)
- governance + audit for agents
- actions as an API, not buttons

Agents don’t “use Salesforce”.
They call it.
And they’ll only call something that behaves like AWS.

[Aurelien Aubert @aureeaubert]

AI is killing UI-first SaaS.

Not slowly.
One category at a time.

Watch YC batches.
Every 6 months, a new batch ships products in the same category:
cheaper, faster, and more AI-native.

Not because founders are copying.

Because once a product lives at the UI layer,
iteration speed kills differentiation.

Most SaaS was built for humans clicking buttons.

AI doesn’t click.
It doesn’t navigate UIs.
It doesn’t follow your “best-practice workflows.”

It calls systems.

That’s why apps are getting fragile.
And infrastructure isn’t.

Every durable layer looks the same:
compute → @awscloud
payments → @stripe
auth → @WorkOS

GTM is going the same way.

AI doesn’t want another tool.
It wants primitives it can rely on.

Build an app if you want distribution.
Build infrastructure if you want to exist in 10 years.

[Aurelien Aubert @aureeaubert]

@mike_matas Wow

[Aurelien Aubert @aureeaubert]

GTM engineers… you need to wake up.
If you are building Clay tables, you have nothing of an engineer.
And you will be replaced in the next 6 months.

A real engineer doesn’t build notebooks.
They build systems.

Real GTM engineering is uncomfortable:
- clear data models instead of spreadsheets
- deterministic pipelines instead of click paths
- systems that retry safely and don’t duplicate data
- explicit schemas, versioning, and guarantees when things change
- outputs sales can trust, not dashboards they question

So stop trying to be a Clay expert.
That path ends with you hitting a wall.

[Aurelien Aubert @aureeaubert]

AI is flipping build vs buy.

You’ll build more (code is cheap).
But you’ll buy more infrastructure (running it is hard).

Because production isn’t prompts, it’s:
- connectors that don’t break
- data that stays consistent
- workflows that are reliable
- governance you can trust

The next generation of GTM will be built like engineering:
open platform + primitives + APIs.

So we’re making Cargo APIs public:
build your own models, sync Salesforce, query, enrich, orchestrate.

Build whatever you want.
Just don’t rebuild the plumbing.

https://t.co/CcfiDUy0ZH

[Cargo (YC S23) @cargo_hq]

Feature - Configure retry policy to instruct a workflow how to retry upon a node's failure

[Y Combinator @ycombinator]

Welcome to YC, @aureeaubert, @Maxlogike, and @cargo_hq!

Cargo (YC S23) is the revenue architecture built for modern teams. Easily segment, score, and route leads to turn pipeline into revenue.

https://t.co/Kc7aChwXHO

[Aurelien Aubert @aureeaubert]

@JuneDotSo 😂

[Aurelien Aubert @aureeaubert]

@0zne Huuuge, definitely going to use it @cargo_hq

[Cargo (YC S23) @cargo_hq]

We are listed on @ProductHunt 🔥

Try our “AI for Sheets” addon and support us 🙏🏻

https://t.co/LGhxPkBstp

[Cargo (YC S23) @cargo_hq]

Refining our text editor ✍️

[@levelsio @levelsio]

DUDE WTF https://t.co/RzEfPl7r6i