Very excited to share voice calls for AI agents on Plivo now ₹0.19 /30 seconds in India.
Happy Independence Day 🇮🇳
6 billion calls a year. 5,000+ builders across India. One consistent ask: lower pricing.
Done. India domestic calls via Plivo Voice APIs or SIP trunking now cost ₹0.19 per 30 seconds.
Get started here - cx.plivo.com/signup
Happy Independence Day!
Open source and self hosted claude artifacts
github.com/plivo-labs/glance
Heard today from two sources that @AnthropicAI maybe looking to buy wrap to own the console as Claude code gains traction.
Find it very odd that the Open AI team doesn't use ChatGPT Atlas as their default browser! 🤔🤔
Which browser do they use by default - Chrome/Comet?
OpenAI's Atlas vs Perplexity's Comet - Whose better??
While both are chromium based and can perform agentic tasks, after real world testing, there's a clear winner for now : @perplexity_ai @comet 🏆
Perplexity's Comet is more reliable for daily tasks, especially for page monitoring, always on agent mode, and faster execution. Atlas has better design and feels super fast but needs work on core functionality.
- Design → Atlas: Clean Mac-native feel | Comet: Functional but can look less polished
- Promo Codes → Comet: Faster, stops after finding the best deal | Atlas: Works but slower, keeps trying unnecessarily
- Video Timestamps → Comet: Quick YouTube navigation to exact moments | Atlas: Eventually works but much slower
- Website Navigation → Both handle Airbnb/Zillow well | Atlas switches sites on its own to compare, Comet sticks to original
- Page Monitoring → Comet: Can refresh indefinitely & take action | Atlas: Refuses simple refresh tasks ❌
- Agent Mode → Comet: Always-on automation | Atlas: Manual activation needed each time (annoying)
- LinkedIn Search → Comet: Finds actual profiles & starts messages | Atlas: Just suggests celebrity names
- Tab Management → Comet: Auto-opens multiple tabs | Atlas: Requires manual clicking
Watch the full video here - youtube.com/watch?v=g_lBZ-EJ…
Insane I missed this earlier - "How Elon Works"
New episode: "How Elon Works"
This episode covers the insanely valuable company-building principles of Elon Musk
A few notes from the episode:
1. The mission comes first.
2. Retreat is not an option.
3. A maniacal sense of urgency is our operating principle.
4. Product design should be driven by engineers.
5. You should not separate engineering from product design.
6. Having separate design and production departments is bullshit. Keep everything together and feedback immediate.
7. The leader should be on the front lines. You should be a battlefield general.
8. "If they see the general out on the battlefield, the troops are going to be motivated. Wherever Napoleon was, that's where his armies would do best."
9. Apply The Algorithm constantly. (1) Question every requirement. (2) Delete any part of the process you can. (3) Simplify and optimize. (4) Accelerate cycle time. (5) Automate.
10. Repetition is persuasive. "I became a broken record on the algorithm. I think it's helpful to say it to an annoying degree."
11. You should go ultra-hardcore on deletion and simplification.
12. Camaraderie is dangerous. It makes it hard for people to challenge each other’s work. (Refer to point #1)
13. Never ask your troops to do something you wouldn’t do.
14. Hire for attitude. Skills can be taught. Attitude changes require a brain transplant.
15. Good attitude = A desire to work maniacally hard.
16. The only rules are the ones dictated by the laws of physics. Everything else is a recommendation.
17. Keep your entire company committed to a common goal.
18. If things aren’t going well, throw away the existing design, start from first principles, question every requirement based on fundamental physics.
19. Find the limit. You want to delete as much as possible and you can’t do that unless you find the limit.
20. If you aren’t adding back at least 10% of the things you deleted, then you didn’t delete enough.
21. Maintain control. Avoid joint ventures. Eliminate middlemen.
22. Have a relentless dedication to questioning every requirement.
23. No work about work, just work.
24. Go to the problem. Get on the plane. Fly to the source. Go to the exact location in the factory. Go to the problem and stay there until it's resolved.
25. The best part is no part.
26. Be wired for war.
27. Do not fear losing. It hurts the first 50 times but then you’ll be able to play with less emotion. You will take more risks.
28. Stay heads down focused on doing useful things for civilization.
29. When something is important and has to be done quickly, have meetings every 24 hours to run the algorithm and check on the previous days progress. You'll be shocked at how fast this speeds things up.
30. Life needs to be interesting and edgy.
31. Delete, delete, delete, delete.
There are 100 more ideas in the episode. I hope you listen to it. 30 years of Elon’s career + 60 hours of reading and research and me just absolutely ripping through idea after idea at 2x speed for 90 minutes.
It will be hard to find a better use of time.
Crazy!! 🔥🤪
Polymarket shows a 74% chance on Google have the best LLM model by end of 2025. Maybe Gemini 3 coming soon?
For all Claude lovers, here is your chance to make/lose some money if you think otherwise.
Just tried OpenAI Atlas.
First impressions, its insanely fast. Just like how superhuman was for email. 👇👇
A new kind of browser it is from OpenAI - Atlas!
Built on chromium, Atlas has everything a regular browser has and looks familiar to chrome, plus also has:
- ChatGPT right within Atlas
- Browser memory within Atlas that works with ChatGPT to personalize experiences
- Action based agents, that can work better than operator
- Can search web history in natural language
-
This is going to be fun - Comet from perplexity, Atlas from OpenAI and Chrome from Google. Are more browsers coming?
A new kind of browser it is from OpenAI - Atlas!
Built on chromium, Atlas has everything a regular browser has and looks familiar to chrome, plus also has:
- ChatGPT right within Atlas
- Browser memory within Atlas that works with ChatGPT to personalize experiences
- Action based agents, that can work better than operator
- Can search web history in natural language
-
This is going to be fun - Comet from perplexity, Atlas from OpenAI and Chrome from Google. Are more browsers coming?
What is OpenAI going to be launching? 👀👀
1. Browser named Aura
2. Healthcare related
3. Something for chrome
These are the options from the grapevine!!
Which one do you think it is in the next hour?
Deepseek-OCR model from @deepseek_ai is a total game changer!
Compresses 7000 text tokens → 800 vision tokens
(10x compression) at 97% accuracy.
Traditional attention needs 49M operations (7000²) vs them 640K (800²), that's 77x fewer operations!
And it's multilingual - supports 100+ languages.
Why this matters:
- Extends context windows 10x (fit entire books in just 128K context!)
- Enables unlimited context via memory fading (like human memory)
- Could make RAG redundant for many use cases
50-70% cheaper than AWS/Google/Azure OCR
- Imagine this for video/image analysis, robotics, autonomous driving 🤯
Meta, OpenAI, Google, Anthropic, Grok - you all need to add this ASAP!
github.com/deepseek-ai/DeepS…
Woah - what the fish! Text to speech that sounds so real. Watch out Elevenlabs.
I tried the elon musk and the donald trump voices. Hilarious
Try here: fish.audio/app/text-to-speec…
AI has re-ignited the PLG conversation. I'm hearing more and more founders want their products to be PLG today. Almost everyone feels AI can help customers see instant results, making PLG motions more efficient.
Companies like @cursor_ai @Replit have shown that getting to $100M can be done very fast with just PLG, but is it also because they serve technical personas?
The risk most people highlight is shallow adoption - lots of signups that don't translate into paying customers or high churn. Maybe the earlier approach works better - to have the product be super intuitive with product-led entry points (free trial, self-serve) and have human-led expansion with sales/account managers only when the customer wants it!
Wild analysis from WaPo on AI: 🤯🤯
- Last year, the best open AI models were American. This year? Every single top model on LM Arena is Chinese.
- And on Hugging Face, Chinese models like DeepSeek & Qwen are getting 2× more downloads than U.S. ones.
The gap is widening!
Elon just announced Grokipedia v0.1 launching on monday. The goal is interesting - its not to just list facts, it’s to understand them.
Most AI LLMs trains on Wikipedia. Garbage in, garbage out. If this works as promised, this could be a game changer. Look out Wikipedia!
Replying to @Polymarket
Buggy beta version of Grokipedia V0.1 will be released on Monday.
Even this very early release is better on average than Wikipedia imo.
We will be looking for as much critical feedback as possible.
And we will offer you the opportunity to donate $5 to send a Grok dick pic to Jimmy Wales.












