Miguel Sanchez

I turn ideas into finished creative with AI.
Building the workflow from idea to execution.
Sharing what works, fails, and costs.

[Miguel Sanchez @MSanchezWorld]

I was lucky to be young when the web arrived.

I started as an animator, obsessing over keyframes, motion and pixels. Then Macromedia Flash, later Adobe Flash, showed me that animation could be controlled by code, so I became a craftsman of code too.

For years, my advantage was being good at both. I could make things look good and make them work. That skill got me jobs, freelance work, helped me build a company and eventually work with some of the biggest brands in the world.

Now I’m watching both crafts disappear into something new.

I don’t think most people will look at lines of code much longer. You’ll tell an AI what you want, see what it makes, and keep directing it until it works exactly how you imagined.

I think the same thing is coming for video.

We romanticize timelines, keyframes, layers, mice and keyboards because those are the tools we spent decades mastering.

But the timeline was never the goal.

The code was never the goal.

The goal was making something exist that didn’t exist before.

When I first learned animation, I wanted to create my own characters and universes. I assumed I’d have to spend my career working my way through the industry and maybe, one day, earn the opportunity to make them.

That isn’t true anymore.

If you can imagine something, you can increasingly just describe it, see it, react to it, and keep directing until it becomes what you had in your head.

I understand why that scares people. I spent decades learning skills that AI is rapidly making less necessary.

But I think there’s another way to look at it.

Think about everything you’ve wanted to create in your life but didn’t have the money, skills, team or time to make.

How much of it can exist now?

That might be the greatest creative opportunity in history.

Enjoy the ride.

[Jensen Huang @JensenHuang]

For my first post, I’m sharing a letter @NVIDIA signed on why open models matter.

AI will transform every industry, power every company, and be built by every country.

Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty.

The world needs both frontier closed models and frontier open models.

https://t.co/AUKzoQ5Ikb

[Nikita Bier @nikitabier]

The risk with hiring a domain expert for a zero-to-one product is that new products that breakout are always outliers and break the rules.

Foolish people who start from a clean slate and operate from first principles don't have that scar tissue that can cloud decisionmaking. For example, Airbnb had no hoteliers on the founding team.

However, the best scenario seems to be composing the team with both types of people, as it will create the chemistry to avoid obvious mistakes while also questioning "expert" assumptions.

[Miguel Sanchez @MSanchezWorld]

Is all of business now us all competing to sell to people who don't know how to use Claude or ChatGPT?

[Idea Browser @ideabrowser]

Fable 5.2 on the business moats for next 10 years

[Miguel Sanchez @MSanchezWorld]

The only real option is to build systems that need multiple models, since frontier labs won't promote their competitors' models.

[Matthew Berman @MatthewBerman]

I keep trying to warn you.

If you build software, the labs will compete with you.

[Miguel Sanchez @MSanchezWorld]

@AlastairGrayson Ai has to write the safe code now

[Miguel Sanchez @MSanchezWorld]

Insane that in just a few years all human code has become technical debt and a liability now.

[Peter Steinberger @steipete]

"When the company ran diagnostics on about 700 systems, it found 10,500 vulnerabilities and held an emergency press conference on June 16, applying fixes over three months.
However, when it re-ran diagnostics using "GPT-6 Astra," the latest model released by US-based OpenAI, it confirmed 35,000 new vulnerabilities. "Honestly, we're completely exhausted" https://t.co/0T069oG6Zh

[Miguel Sanchez @MSanchezWorld]

Decent advice

[klöss @kloss_xyz]

Fable 5.2’s advice for the next 10 years.

[Miguel Sanchez @MSanchezWorld]

@MysterONYX @DaveShapi Bingoooo

[Miguel Sanchez @MSanchezWorld]

Damnnn this is so true

[Danny Postma @dannypostma]

Marketing cofounder is the new technical cofounder

[Miguel Sanchez @MSanchezWorld]

Isn't it a running joke that most of Silicon Valley is in the same situation as Jensen Huang? It's sad that after working so many years, this is what he'll be remembered for. I hope that doesn't happen.

[@levelsio @levelsio]

It's never been this easy to build software

And never this hard to make money with it

[Miguel Sanchez @MSanchezWorld]

Pretty crazy how fast this happened.

[Sanabi @0xSanabi]

This is f*cking insane. Boris Cherny, the man who built Claude Code, was asked if people should still learn to code. His answer:

"In a year or two, it's not going to matter"

In 73 seconds, he explains what to do instead: play with the AI tools, learn them, don't be scared of them, stay on the frontier

On his team, everyone codes now. The PM, the designer, even the finance guy

He hasn't edited a single line of code by hand since last November. He ships 10 to 30 pull requests a day

Writing Code → Prompting → Agents → Harness → Loops → Everyone on the Team Ships

Claude Code is a harness: the tools, memory, rules and loops around the model. His team changes it with every new model and cut 80% of its system prompt

Save this before it disappears from your feed

Every doc, guide and repo below is free:

Claude Code best practices
https://t.co/GbmXiaEIDd

Effective harnesses for long-running agents
https://t.co/xrtGydQVJY

Building effective agents
https://t.co/AlmLkMTGu2

Claude Code docs
https://t.co/HD2f3EEFmE

Memory: how Claude remembers your project
https://t.co/nozmp2WL3o

Custom subagents
https://t.co/VKlbDuol0B

Hooks: rules the agent can't skip
https://t.co/qsDseCAMX5

Skills
https://t.co/5bEJ0arUGo

Run Claude Code on a schedule
https://t.co/JolCHBKW2p

Free course: Claude Code in Action
https://t.co/Dd0MnjEGSn

Claude Code on GitHub
https://t.co/PLMi4HwjFu

Claude Code inside your PRs and issues
https://t.co/YuA7O7ENw2

Build your own agent on the same harness
https://t.co/JidFNfxArL

The best Claude Code resources in one list
https://t.co/BeOsrePwHY

Plugins, agents and skills for Claude Code
https://t.co/snrWwfOncP

He compares it to the printing press. Writing went from a few scribes to everyone. Building software is next

95% will keep asking whether to learn to code, the 5% who build a harness will ship 30 pull requests a day

Bookmark it and read the first link tonight

Then use the guide below to build your harness before everyone else does ↓

[revayz @0xrevayz]

Dario Amodei (Anthropic CEO) just released a prompt that turns Opus 5.5 into a dev team that gets smarter every night you sleep

It's f*cking unreal...

Send it to Opus 5.5, and thank me later. Then read the article below ↓

[Shann³ @shannholmberg]

now is the best time to learn marketing engineering (how to become a cracked marketer)

[0xMarioNawfal @RoundtableSpace]

DeepMind CEO Demis Hassabis:

"One person who really understands AI will outperform an entire startup team

In 5 years, the people who build with AI will run whole companies with a tenth of the headcount"

[Miguel Sanchez @MSanchezWorld]

Just in time for those data centers to get killed. SMH.

I once asked a Democratic politician why so many Democrats opposed crypto when it could help the people they’re meant to support.

They said, “Because they don’t understand it.”

I asked, “Isn’t that their job?”

They noted NYC depends on financial industry tax revenue—around 30%.

That stuck with me.

For years, people said false things about crypto. When questioned, they hadn’t researched it—just repeating hearsay.

Same with AI now.

People criticize data centers’ water use. Golf courses use plenty too, and they’re everywhere—why no similar outcry?

Most don’t realize how long it takes to understand these technologies: not just using them, but their potential, who they help, and real risks.

I’ve studied crypto for years and still learn. Same with AI. Deeper understanding reveals more possibilities.

Is it realistic to expect politicians to grasp every technology, industry, and outcome? No.

Yet we rely on a system where those who may not fully understand decide their futures.

Politics worsens it: sides form based on who promotes it, not what it does.

Imagine rejecting computers or the internet over which party benefited.

We should question technologies, but need better ways to understand and decide—especially ones that could transform millions of lives, including those needing opportunities most.

We’re fighting over today’s beneficiaries instead of understanding tomorrow’s. This extends beyond crypto or AI.

[Miguel Sanchez @MSanchezWorld]

Horses were once everywhere because humans needed them. Once cars arrived, their population declined, and many people never see a horse in person.

We have as many humans as we do because they were needed for the economy. What happens when they're no longer needed?

I hope we don't go the way of the horse.

[Miguel Sanchez @MSanchezWorld]

@DaveShapi If I don't end up with a good relationship with my children I would feel like a failure but that's just me.

[Miguel Sanchez @MSanchezWorld]

@MiaAI_lab Chinese models possibly with backdoors?