dggtn

🇪🇨 Developer living in 🇦🇷 |shefi cohort 13 |

[sophia dew @sophiadew]

vincent weisser might have one of the best goodreads lists of all time. great mix of philosophy, science, technology, history, spirituality and practical wisdom

[Bober_smart @Bober_smart]

It is becoming f*cking dangerous: I don't understand why everyone still isn't doing this, thanks to this I reached an income of $12,000 as early as 3 months ago

Andrej Karpathy, co-founder of OpenAI, shared a straightforward concept that blew up with over 23 million views: stop relying on AI strictly for coding and start using it to construct your own personal knowledge engine

Point Claude Code to a folder, toss in raw materials (whether it's an article, a video transcript, or a PDF) and Claude processes everything, links relevant topics together, and organizes it into a self-updating wiki of your personal knowledge. It acts like compound interest: the more information you add, the more capable it becomes

Here is the exact breakdown:

Set up Obsidian, create a fresh vault, and open it using Claude Code

Supply Karpathy's wiki layout prompt and instruct Claude to build the architecture

Claude automatically sets up three components: a ⁠raw⁠ directory for source files, a ⁠wiki⁠ directory for processed pages, and a ⁠CLAUDE.md⁠ control file to run the system

Dump any document into the ⁠raw⁠ folder and run "ingest this"

Query your entire knowledge repository at any time, indefinitely

It takes just five minutes to configure, eliminating the need to ever start a conversation with a completely blank context window again

Check the link below for the complete step-by-step walkthrough on setting up Claude with Obsidian

Bookmark this post for later

[shobhanai @HeyShobhan]

They built an AI agent in 20 minutes and sold it to Anthropic for $100k - and at Stanford showed how to repeat it from scratch:

02:00 - how to build your first agent with Opus 5 in 20 minutes
22:34 - one agent replaces an Anthropic engineering team worth $200k a year
49:47 - how to make $1.2M a year selling agents to Anthropic and Google

after watching I spent 20 minutes in Claude Code - the agent replaced 10 Anthropic engineers and they paid $100k and asked for more.

save & watch - the article below shows how to build an agent in Claude Code in 20 minutes and reach $1.2M a year.

[naldz. @naldzyul]

Esta función de Claude Code es una auténtica locura 🤯

Este chaval te enseña a crear un equipo de agentes de IA en solo 1 minuto.

[robin. @robinstetic]

este tío acaba de contar cómo usa claude code + youtube para sacar $97k al mes

son menos de 15 minutos y merecen la pena, guárdalo para luego🙇🏻‍♂️

[Moo Jin Kim @moo_jin_kim]

What a stunning launch by @sundayrobotics. The results are super impressive: long-horizon & whole-body control spanning multiple minutes, generalization to unseen environments, and impressive dexterity — all with ZERO robot data! 💯 So excited for this team's future.

[Moo Jin Kim @moo_jin_kim]

We release Cosmos Policy 💫: a state-of-the-art robot policy built on a video diffusion model backbone.

- policy + world model + value function — in 1 model
- no architectural changes to the base video model
- SOTA in LIBERO (98.5%), RoboCasa (67.1%), & ALOHA tasks (93.6%)

🧵👇

[Gastón Mira @gastonmdq]

Estos son algunos de los que sigo, no fallan… tanto contenido en inglés como en español

Global 🌎
@karpathy: investigador, ex OpenAI
@sama: CEO de OpenAI
@gdb: cofundador de OpenAI
@DarioAmodei: CEO de Anthropic
@demishassabis: cofundador de DeepMind
@rauchg: CEO Vercel
@AndrewYNg: educador de IA
@simonw: blogger de LLMs
@bcherny: creador de Claude Code
@trq212: Claude Code
@addyosmani: Claude Code
@ClaudeDevs: novedades de Claude para devs
@OpenAIDevs: novedades de OpenAI para devs
@thsottiaux: Codex en OpenAI
@steipete: creador de OpenClaw
@poteto: Grok Bot en SpaceXAI
@mattyp: Grok Bot en SpaceXAI
@theo: CEO de T3 Chat
@rileybrown: builder de agentes
@mattpocockuk: educador dev
@vercel_dev: comunidad dev Vercel
@dhh: no hace falta presentación

🧉Latam / contenido en español.
@santtiagom_: lo mejor de Latam en AI
@midudev: programación y AI
@MoureDev: programación y AI
@G_Programming: comunidad Gentleman Programming
@AlanDaitch: AI sin humo
@antonioleivag: builder con agentes
@barckcode: modelos abiertos
@tebayoso: product dev AI
@ayebytes: bitcoin y AI
@AleRepetto5: AI engineer
@LuigiMdpDev: embajador SpaceXAI
@juampitech: growth en Firecrawl
@leandrojmora: IA y deporte

Si me olvido de alguien (segurísimo), les pido disculpas… a quién sumarían?

[Dhruv kumar @Dhruvkumar16797]

Anthropic engineer: "at Anthropic, we don't write prompts anymore. we build loops"

in 42 minutes, she shows how the Claude team creates loops that prompt themselves

if this cost $400, people would call it the best agents course of the year

it's free

watch it,

[Ryan Carter @Grow_withAI]

One of my friends makes $1.2M a year as an Anthropic engineer.

I asked him how he became so good at prompting. He sent me a video that was never meant to be public: Anthropic’s core team prompting playbook.

You’ll struggle to find a better 30-minute guide to prompting than this.

I watched it last night. Halfway through, I realized I’d been using Claude completely wrong for the past two years.

Watch it. Bookmark it.

[lis @lisaknowsai]

SIGAN A ESTA PERSONA NO DIGO NADA MÁS…

[Stratis Tsirtsis @stratis_]

If you want to do a PhD in machine learning, in Europe, on topics that are conceptually novel and technically challenging, check this out. Almost 7 years ago, I moved to a new country to start my PhD with Manuel, and it was one of the best decisions I've made 🙂 Apply.

[Suraj Sharma @suraj_sharma14]

Do yourself a favor!

– Learn AI Engineering
– Build an agent that fixes one painful, boring workflow
– Give it tools, context, memory and MCP
– Write evals before you write prompts
– Deploy it with a live link anyone can try
– Route easy requests to small models, hard ones to reasoning models
– Cap every run with token, time and cost budgets
– Cache repeat queries so they cost almost nothing
– Retry failed tool calls with backoff not blind loops
– Validate every tool output against a strict schema
– Make agent steps resumable so crashes don't restart everything
– Run every tool call in a sandbox with least-privilege access
– Strip secrets and PII before anything reaches the model
– Treat every web page and document as possible prompt injection
– Require human approval before money, emails or deletes
– Log every model call, tool call and decision with traces
– Score the agent's path, not just its final answer
– Build a golden test set from real user failures
– Block deploys in CI when eval scores drop
– A/B test new models on a slice of real traffic first
– Track cost per task not cost per token
– Stream responses so users never stare at a spinner
– Turn every thumbs-down into a new eval case
– Fine-tune small models on your best production outputs
– Add a fallback model for when your main provider goes down
– Build a browser fallback for apps with no API
– Version your prompts like code
– Write a README that explains the architecture in 5 minutes
– Record a 60-second demo of it working
– Share the code on GitHub and add it to your resume

You won't believe how many job offers you'll get.

Bookmark this list. Build it all.

[Dylan Txa @Dylan_txa_]

I’m 31. I built @GojiberryAI to nearly $5M in ARR and got accepted into Y Combinator.

If I had to start from 0, here's exactly what I'd do:

1. Sell it before I build it.

No code. Just a simple slide deck (mine was 6 ugly slides) explaining the problem, the solution, the outcome, and the price. I made my first $10k that way, before writing a single line.

2. Pick a painfully specific customer.

Not "B2B SaaS." Something like "founders at 20-person SaaS companies about to hire their first SDR." So specific that the right person reads it and thinks "that's me."

3. Start outbound on day one, but only to people showing intent.

Not scraped lists. People engaging with competitors, changing roles, raising money, or publicly posting about the exact problem I solve. That's the gap between a 1-2% reply rate and 25-40%.

4. Lead with value, never a calendar link.

Send a blueprint, not "got 15 minutes?" Let the resource do the selling, and the trial becomes the obvious next step instead of a pitch.

5. Pick ONE channel and go deep.

For us it was outbound first, then Reddit (10M+ organic views), then LinkedIn lead magnets. I wouldn't touch a second channel until the first one was clearly working.

6. Talk to customers every single day.

The product doesn't matter until you understand the problem better than they do. Spend 90% of every early call listening, not demoing.

7. Only build once people are actually paying.

Then keep it dead simple and price it to sell itself. We landed on $99/mo with a free trial, so the funnel runs without me dragging anyone onto a call.

8. Do this relentlessly for about 14 months.

That’s roughly how long $0 to $5M took us.

Bootstrapped.

No outside funding.

Most founders don’t lose because they can’t build.

They lose because they build too early, sell too late, and quit the channel before it compounds.

[AI_Explorer @ai_explorer25]

Best accounts to follow from each frontier lab to stay constantly up to date

Anthropic
@karpathy -must-follow account for AI; recently joined Anthropic
@bcherny- Claude Code creator, always shares great tips
@trq212- also a Claude Code developer; writes amazing articles on CC

OpenAI
@polynoamial- works on reasoning research, shares a lot of technical details
@gabriel1 - Sora developer, great career path
@jxnlco - works on dev experience, shares a lot about Codex

Google AI
@OfficialLoganK- all the major Google Gemini and AI Studio updates
@ammaar - product and design; shares great things about vibe-coding in Google AI Studio
@fofrAI - cool use cases for generative models

Cursor
@leerob- the loudest voice behind Cursor updates
@ericzakariasson- shares great insights on using Cursor
@mntruell - Cursor’s CEO; major releases and usage updates

xAI
@ai_explorer25- ex-MSFT, covers all ai content and free resources
@milichab - recently joined xAI, shares updates on Grok
@skcd42 - also covers major Grok releases

[Azka Ai @warraich_Ai2]

Anthropic senior engineer just shared a 1-hour course on building a team of agents with loops & graphs:

00:27 - claude-md & Plan mode
11:24 - building skills and hooks
37:02 - agents and subagents with Claude
52:47 - self-improving loops and graphs

taught by someone who runs this every day inside Anthropic

Prompts → Agents → Loops → Graphs

a loop closes one job without you

a graph decides which jobs exist and carries every accepted one into the next run

watch it today

then save the full guide on building a team of self-improving agents below before everyone catches up ↓

[Codie Sanchez @Codie_Sanchez]

This guy coaches Sam Altman.

He also works with leaders at Anthropic, Google, Apple, and SpaceX.

I sat down with him to learn how he coaches billion dollar businesses.

Here are Joe Hudson’s 7 operating rules:

1. Kill your 1:1s
These meetings lead individuals to solve problems instead of the team. Much better to bring problems to the whole team so everyone feels responsible for solving them. I agree, haven’t done them for years...

2. Rotate who runs recurring meetings
When the same person always leads, people focus on pleasing or rebelling against them. Spreading the authority makes everyone accountable to the team instead.

3. Give anyone the power to change a bad meeting
Everyone already knows when a meeting isn’t working (and usually why). Giving people permission to say it makes meetings more effective.

4. Make every employee run at least 5 experiments a quarter on their own job
It makes the team more productive, keeps people interested, and helps retain smart employees who would otherwise get bored.

5. Ask your team to push back on your ideas
If you try to sell them on the idea, they may nod in the room and do nothing afterward. Asking for objections tests the idea + gets everyone aligned before you commit to it.

6. Start by building the part you understand the least
If you leave the confusing part until the end you might find out it doesn’t fit and have to rebuild everything around it.

7. End every email with an action needed
Name the person responsible, the action they need to take, and the deadline. Joe says this rule increased productivity in his business by 35 to 40%.

Thanks for coming on @FU_joehudson and sharing how you run your business.

Search “BigDeal pod” on youtube.

[S 🫧 @He1s_Sammy]

THIS IS F**KING INSANE

THIS 23 YEAR OLD USED CLAUDE TO REVERSE ENGINEER A KIDS CHANNEL WITH 28.5M SUBSCRIBERS.

then reportedly made $10K from YouTube Shorts.

he didn’t copy individual videos.

he studied the formula behind them:

→ recurring characters
→ simple stories
→ bright visuals
→ fast pacing
→ formats already proven to generate massive views

then he built an AI production engine around those patterns.

Claude analyzes the channel, identifies repeatable ideas and writes original scripts.

connected AI tools handle the rest:

→ characters
→ voices
→ music
→ animation
→ final assembly

his involvement?

almost nothing.

he chooses the direction, approves the strongest concepts, and lets the system handle the repetitive production.

one prompt → one finished Short.

and because the workflow is reusable, he can keep producing without rebuilding everything from scratch.

the original channel reached 28.5M subscribers across 798 uploads.

he didn’t copy one cartoon.

he reverse engineered the machine that could produce an entire channel.

that’s the real AI advantage.

[nerea @nereasolenne]

GOOGLE HA SOLTADO ALGO QUE NADIE ESPERABA.

Un tutorial de 8 minutos sobre cómo construir tus primeros agentes de IA.

[draven @dravenip]

Anthropic just launched a free 4-hour course.

It teaches you to use Claude with the same workflows their own engineers use internally.

00:15 → How to structure prompts for Claude the right way
11:41 → The "output contract" that fixes inconsistent responses
30:04 → The loop that makes Claude's agents more reliable
33:21 → Why Claude struggles with certain coding tasks
1:33:39 → The workflow Anthropic engineers use every day
2:50:56 → The simple tweak that makes Claude noticeably smarter

No Anthropic account, no API key, no subscription needed. Just an email on Skilljar.

When you finish, you get a certificate signed by Anthropic. There are already several free courses total, and most "AI experts" on LinkedIn never took any of them.

Save this and watch it today.

Then read the article below.