Pratik

Loves Techno-law-gy & Techno-logical stuffs
#Okta #IAM #CyberSecurity #identity #ai-iam #claude-paglu
Opinions are my own

[Pratik @pmbhatt]

"Facts do not care about your feelings"
- Ben Shapiro

[Nix0n @UnTalNixon_exe]

🚨EXCLUSIVA: CHINA ACABA DE ABRIR LA PUERTA AL FUTURO DE LAS WEB APPS

Alibaba open-sourced PageAgent.

Es un agente de IA que vive dentro de tu página web.
Con una sola línea de código, conviertes cualquier web en una aplicación controlada por lenguaje natural.

Ejemplo real:

- “Click login, fill my credentials and submit the form”
- “Go to the dashboard, find the Quick Start section and summarize it in markdown”

Todo esto sin extensiones, sin backend, sin capturas de pantalla y sin reescribir nada.

Solo agregas esto:

```html
<script src="CDN_URL" crossorigin="true"></script>

Esto no es una demo. Es producción-ready.

REPOO 👇👇

[Pratik @pmbhatt]

@dabit3 @cognition @DevinAI Applied

[Pratik @pmbhatt]

Chief Economic Advisor's Warning: Will Cash Flow Change Your Career? | D... https://t.co/LnsH4duLH9 via @YouTube

[Amish Thaker @amishthaker_]

Kudos to @kunalb11 for bringing $900 M FDI into India. 😎🤘🏼

[Amish Thaker @amishthaker_]

Great to see an Indian chocolate brand. Shoutout to @prajnay_darkins and team.

Available on Zepto and Blinkit.

https://t.co/Wy9f4Ynz31

[︎ ︎venom @venom1s]

She saw her mother cooking twice and called it slavery.

But,

She didn't see her father

- Coming from small towns and villages to the city
- Working hard to buy land, build a house
- Provide a good education and life for his kids
- Never taking a day off
- Sacrificing all his dreams and his entire life for their well-being.

But no man will call it slavery or oppression.

We see it as our duty. All fathers deserve respect.

[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.

[Pratik @pmbhatt]

@jaanvii0 there is no company called https://t.co/hhA6ElFy2C - it takes you to microsoft sign-in page. Share your linkedin profile if actually there is a company where you work.

[Amish Thaker @amishthaker_]

Happiness is when clients are happy enough to post feedback publicly !!

Thank you sir ji for your trust and support 😀🙏 !!

@TheBigGeek

[Pratik @pmbhatt]

@i_m_prapti TCS Nasik - ગુજરાતની એક પણ સ્ત્રી ઈનફ્લુએન્સર આ વિષે ના બોલી - જાહેર માં એમને ટેગ કરવા છતાં. એમની માટે ધુરંધર નો છેલ્લો સીન જેમાં એ મેજર ઇકબાલ ને કહે છે, એ જ મને કહેવા નું મન થાય છે. 😁

[codila @0xCodila]

Anthropic Quant Andrej Karpathy:

"Most people use tools that they don't understand

- the ones who strip everything down to basics - end up faster than everyone else "

"the best code is the code anyone can read "

he couldn't fix a bug in 2 hours, so instead of googling - he rewrote the entire system from scratch

no frameworks. no dependencies. it ended up faster

that's the difference between using AI and understanding it

25-min masterclass - bookmark and watch

[The Emissary @TheEmissaryCo]

How a single Indian anon (@AgniVesa_07) infiltrated &amp; exposed 42 anti-Indian hate groups...

...And in the process discovered the disturbing role of an Indian political party in all of this.

[Dr. Shah @ankitatIIMA]

💥Translation -

The dollar printing machine will not be able to print endlessly aka English me bakwaas karne ka aur typing karne ka kuch nahi milega 😅

[mememandir @mememandir]

Dear ministry of IT,
Scammers are reportedly using Slack &amp; MS teams to leak NEET papers. Kindly issue a ban.

[Amish Thaker @amishthaker_]

Building a product and building a business are two very different things. I'm learning that the hard way....

[Amish Thaker @amishthaker_]

An AI agent on Whatsapp trained on the business info that keeps answering customers questions is a no brainer. I could not find a solution. So I build one. Turns out it's can actually be a big business.....

[Anand Ranganathan @ARanganathan72]

Spine always grows in the back that is against the wall. America denied us crucial LOX/LH₂ cryogenic engine technology. We developed it indigenously and using it we now send American satellites into space.

India can remain sovereign only if it has a sovereign AI. @narendramodi

[Vasant Shetty | Building Mundhe Banni @vasantshetty81]

India is the second-largest market globally for both ChatGPT and Anthropic. If they can turn off the access at the press of a button like this, we are absolutely at the mercy of a foreign govt. Geopolitics is getting uglier. Globalisation in the current form is dead. This is a huge wakeup call for India.

[S Gurumurthy @sgurumurthy]

America is doing what Dr Abdul Kalam wanted the Indian govt to do in 1998. After pokharan II blast US-West sanctioned India. Kalam told me (which I have recorded many times) to persuade PM Vajpayee (who I was close to) to ensure the sanctions continue for our tech to develop

[Marie Haynes @Marie_Haynes]

This is really big news. Google introduced the Open Knowledge Format (OKF) - a standardized way to store information in a directory of markdown files. Makes it really easy to make a digital brain that agents can use.

These files can serve as a living wiki. You can give agents the ability to query them or edit them. They can interlink.

Seems to me this could replace Notion or Obsidian. I can think of so many uses for this.

Google's blog post: https://t.co/DqSjg4UpvH

An easier to understand explanation is the SPEC.md file:
https://t.co/A3qSz3Tfas

I gave those two links to Antigravity and asked how we could use it for any of the projects we're working on. It came up with so many ideas. I would imagine Claude Fable 5 would whip up some pretty amazing things based on this system.

Currently creating an OKF library of our pepper garden. It's going to be a fun weekend.