Carlos Oblitas

Amante de la tecnología... apasionado por los números... Construyendo el mañana, hoy...

[Akshay 🚀 @akshay_pachaar]

Web scraping will never be the same.

(100% open-source visual search at scale)

PixelRAG is a retrieval system that skips HTML parsing completely.

Instead of scraping a page into text and embedding chunks, it screenshots the page and retrieves the image. A vision-language model reads the answer straight off the pixels.

Why that matters: parsing is where web RAG quietly loses information.

- A single HTML-to-text parser can drop 40%+ of a page.
- Tables, charts, and layout get flattened or thrown out.
- Swapping parsers alone can move accuracy ~10 points on the same docs.

PixelRAG indexes the page a person actually sees. The team built a visual index of all of Wikipedia, 30M+ screenshots, and it still beats the strongest text RAG baseline by 18.1% on text-only QA.

The repo also ships a Claude Code plugin that gives Claude eyes.

It lets Claude screenshot any URL and read the rendered page instead of scraping the DOM. So you can hand it a live page, an arXiv paper, or your local site and ask what it actually looks like.

One setup script. No MCP server, no backend.

How the pipeline works:

- Renders each document (web, PDF, image) to image tiles.
- Embeds them with Qwen3-VL-Embedding, LoRA fine-tuned on screenshots.
- Builds a FAISS index and serves a search API.

A stronger reader model lifts accuracy with no re-indexing, since the index is just pixels.

Everything is open-source under Apache-2.0.

GitHub repo: https://t.co/qun9TjAdmw

Talking about RAG, I recently wrote an article on a new approach that makes retrieval much more efficient by cutting corpus size by 40x, reducing tokens per query by 3x, and improving vector search relevance by 2.3x.

The article is quoted below.

[0xMarioNawfal @RoundtableSpace]

SOMEONE BUILT KARPATHY’S DREAM AI TOOL IN 48 HOURS

* Graphify turns any folder into a searchable knowledge graph and Obsidian-style wiki
* Lets Claude Code reason over codebases, PDFs, images, and research with huge token savings
* Signals a major shift from raw file reading to structured AI-native knowledge systems

Repo: https://t.co/L0TdPXlKS8

[CyrilXBT @cyrilXBT]

ANTHROPIC JUST DROPPED 13 FREE CLAUDE CERTIFICATIONS AND ALMOST NOBODY IS TALKING ABOUT IT.

Not a YouTube playlist.

Not a third-party course.

Official certifications from the team that built Claude.

Free. Forever.

Here is the full list with links:

START HERE

01. Claude 101 — Learn Claude for everyday work
https://t.co/7y3hN0bL8Q

02. AI Fluency: Frameworks and Foundations
https://t.co/juausxFh7O

03. Introduction to Agent Skills
https://t.co/11ZlK1OaVC

FOR DEVELOPERS

04. Building with the Claude API
https://t.co/aJAciAEw3y

05. Claude Code in Action
https://t.co/c0norD7CU0

06. Intro to Model Context Protocol
https://t.co/iywBhaZn8Z

07. MCP Advanced Topics
https://t.co/y2XQ1snBl9

FOR EDUCATION AND NONPROFITS

08. AI Fluency for Students
09. AI Fluency for Educators
10. Teaching AI Fluency
11. AI Fluency for Nonprofits

FOR ENTERPRISE

12. Claude with Amazon Bedrock
13. Claude with Google Cloud Vertex AI

13 courses. 6 skill levels. 5 audiences. 100% free forever.

The engineers getting hired at $150,000 to $300,000 to work with Claude at the highest level are learning exactly this material.

Anthropic's team just made it available to everyone.

Pro tip: Start with Claude 101 then go straight to Claude Code in Action. That is the fastest path from beginner to builder.

Bookmark this before you pay for another AI course.

Follow @cyrilXBT for every Anthropic resource that compounds your skills the moment it drops.

[elvis @omarsar0]

Been exploring a new way to explore AI research papers to discover deeper insights.

Agents are at the center of it.

So far, I've built this little interactive artifact generator in my orchestrator to visualize things.

This allows me to change views and insights (on-demand) from 100s of papers.

Just scratching the surface here. More to share soon.

[AilaunchX @Ai_Tech_tool]

ANDREJ KARPATHY COULD HAVE CHARGED $2,000 FOR THIS COURSE.

He put it on YouTube.

The full training stack. Tokenization. Neural network internals. Hallucinations. Tool use. Reinforcement learning. RLHF. DeepSeek. AlphaGo.

3 hours of the most comprehensive LLM education that exists anywhere at any price.

Not how to use the tools.

How the entire system was built from the ground up and why it behaves the way it does.

The engineers who understand this build things the ones who only use the tools cannot even conceive of.

The gap between those two groups is not 3 hours.

It is everything those 3 hours quietly unlock for the rest of your career.

[Quant Science @quantscience_]

According to Ray Dalio, the easiest way to adjust for risk is to seek uncorrelated returns.

Ray's made billions from a simple idea.

Here's how to do it in a few lines of Python code:

[DANNY @Danny_Crypton]

🚨 BREAKING: A BIG STORM IS COMING!!!

US Treasury has a massive problem nobody wants to talk about..

Take a good look at this chart.

That giant blue spike?

Yeah… that’s trillions in U.S. debt that expires in 2026. Not 2030. Not 2040.

2026.

And all of it has to be refinanced at much higher interest rates than the near-zero environment it was originally issued in.

In simple terms:

– The U.S. loaded up on cheap debt.
– That cheap debt now has to be rolled over at expensive rates.
– Interest costs are about to explode.
– Something has to give. Markets, taxes, spending, or the dollar.

This is the kind of structural time bomb that doesn’t hit immediately…

but when it does, it hits everything.

Stocks. Bonds. Housing. Crypto.

No market is immune when a sovereign debt wall this big comes due.

Keep your eyes open because most people will notice this after it’s too late.

I was one of the only people who called the top in October, and I’ll do it again, that’s literally my job. Pay close attention.

If you still haven’t followed me, you’ll regret it.

[Quant Science @quantscience_]

How to make a simple algorithmic trading strategy with a 472% return using Python.

A thread. 🧵

[Carlos Oblitas @oblitasss]

@mentorAIhere Guide

[CryptoFace @RealCryptoFace]

7 hours until Bitcoin confirms (an over 80% Bullish indication The Big Boy Bitboy Green Dot on Daily Timeframe.

Even in Bear markets , a Daily Green Bigboy is a very very high probability of a decent size move up.

How long with it last? We will see!

[Crypto Crew University @CryptoCrewU]

IT’S HAPPENING AGAIN

We’ve been waiting for this!

Click Here and watch now: https://t.co/u8HANSIgNk... ⬅️✅

[MMCrypto @MMCrypto]

THIS GIVEAWAY IS STILL ON!

[Carlos Oblitas @oblitasss]

@swissblock__ @bitcoinvector @grok explica si el efecto de esa caida de ayer sera a largo o corto plazo?

[Not Jerome Powell @alifarhat79]

Everyone is waiting for this moment

[Comunidad Andina @ComunidadAndina]

🇧🇴👏🏻 Feliz Bicentenario, Bolivia, País Miembro de la #ComunidadAndina.

Compartimos un mensaje del Secretario General de la Comunidad Andina, Embajador Gonzalo Gutiérrez, en esta fecha especial.

#BicentenarioBolivia #ComunidadAndina #IntegraciónAndina

CC: @MRE_Bolivia

[Fabsverse @fabsverseweb3]

Gaming on Polkadot keeps heating up.

@MoonbeamNetwork is one of the parachains I’ve been tracking the most and has recently upgraded its tokenomics:

→ 100% of GLMR fees are now burned (no more 80/20 split)
→ Treasury gets 80% of bond inflation
→ Staking rewards? Unchanged

What’s the impact? ⬇️

• Less supply = more value for active users
• Ecosystem funding gets smarter
• Long-term design without messing with rewards

Simple change. Big implications.

[naiive @naiivememe]

eth holder go to work today:

[Carlos Oblitas @oblitasss]

Catch the crypto meteor shower from the #BinanceTurns8 celebration! #CryptoMeteorShower https://t.co/HvhT6rUWXw

[naiive @naiivememe]

Bro sold You waiting
for a 2x for 100x