Pinned Tweet
To celebrate the new edition of DDIA, revive my writing discipline and my resolve to read the whole damn thing this time, I share my first blog in years: LSM Trees from First Principles.
ditsuke.com/lsm-trees
there’s a new edition of @intensivedata which is one of the most pivotal technical books written in the last decade, and almost no chatter on this app about the new edition or the new content it covers.
Are there no serious engineers and practitioners on this app anymore? 😢
Tushar retweeted
i cannot stress enough how important it will be to wear a mask when going out from tomorrow, avoid walking outside as much as possible and buying an air purifier for your room.
Tushar retweeted
इस जगह का वीडियो बना कर लौटा ही था कि थोड़ी देर बाद किसी ने बताया कि ऊपर की मंज़िल पर एक फ़्लैट में आग लगी है। आग पर क़ाबू पा लिया गया है। मगर आप लोग AQI की चिंता मत कीजिए। जितना मन करे उतना पटाखा फोड़िए। जिस देश में वोट चोरी से भी किसी का ईमान नहीं हिला उस देश में हवा बचा कर क्या करना।
Can confirm that basically no one wears masks anymore here in Delhi. I was taken aback by this observation since I know more people with respiratory issues here than ever before.
Delhi’s pollution has reached a strange new low.
I’m the only person wearing a mask in my society of 2,500 people. Two elderly doctors have asked me, “Wearing a mask? Everything fine?” I didn’t know how to process it.
We’ve normalised toxic air so much that protecting ourselves from it now needs an explanation.
What a health emergency.
RT @zeeg: i really dont understand why Grok Bot allows you to post to X
just disable it. people are going to do the obvious thing. you kno…
This quoted post is unavailable.
6.1 Sol is **frustrating** to work with. On top of the slow inference, the model rarely takes a straightforward approach, keeps going in loops to waste more time and tokens.
@AnthropicAI @claudeai can you review my appeal? TS-01a1a094-5c0d-70f5-8b0e-b7b005ad15c1 thanks :)
Tushar retweeted
STING OPERATION AGAINST THE WITCH-HUNT OPERATIONS OF LG LADAKH...
For raising my voice for the youth of the nation and the well-being of border areas like Ladakh & Arunachal, I'm being continuously witch-hunted and defamed & our schools being harrassed in every possible way by the LG Administration in Ladakh.
But this Sting Operation by our lawyers establishes that the LG Administration is directly involved in illegal smear campaigns (& much more).
Much more has come to us from the Raj-Niwas/ Lok-Niwas but we're releasing just some excepts for now.
I tried v much to avoid this, thinking it may damage the nation's dignity & morale & hence I first shared the material with the Chief of IB for Intervention, but unfortunately after few exchanges he too stopped responding... So with a heavy heart I am sharing this with the nation as we must stop the LG from playing like an infant in the ecologically & strategically sensitive border regions of India.
@rashtrapatibhvn @PMOIndia @narendramodi @AmitShah @RahulGandhi
Tushar retweeted
Update: A shift in-charge from that same dark store visited me, with a replacement device.
I got on a call with the Zepto SPOC who initiated this replacement, I opened the new Airpods in front of them, this had a different Serial Number on the box, and I was hopeful this would be good.
When I unpacked it, it was a dusty device, when I flipped the lid open, the serial number was exactly the same as the fake device I received yesterday.
Meaning, their entire lot has been compromised.
someone has manufactured multiple devices with the same "fake" serial number.
Depending upon how far the source of this swap goes, looks like many people who ordered Airpods Pro 3 from Zepto (and maybe many other online portals) would have received duplicate products.
Unfortunate but true.
Don't know how deep this nexus goes, but hope the company is able to identify the root cause and fix for good.
Sad that I will not get a Airpods Pro this diwali.
Will probably buy from Bengaluru when I fly back to BLR tomorrow.
xitter.gq/sunnykgupta/status/211…
I hope they find the loophole and fix it too!
I know that the company isn't trying to do this on purpose. But somewhere along the way, there are enough lapses in the process to make room for fraudsters to spawn.
It is a part of almost ever hyperscaling journey.
What is important is how quickly are these found and plugged when identified.
I was only surprised that I explicitly mentioned on the call with customer care that I have identified a security loophole in their Supply Chain and want to talk to someone in Zepto.
The customer support executive was like: "I am the last point of contact, we cannot escalate further, please talk to the Brand".
I mean, what is Apple going to do if Zepto's SCM is broken.
Nonetheless, at least now they have taken notice and will investigate this to its depths.
Tushar retweeted
We've been early adopters of Rust since 2018 and a lot of our infra, voice tech, and even some of the current client is already Rust, the holy grail was getting the whole client there. That means Electron goes away, and so does React Native, mobile keeps native UI but shares app code with desktop via Rust. We hope to get it into testing with users soon™. (And it's not Tauri, it's GPUI, so good guess everyone.)
Alright, back to staring at the WoW Forever login screen.
THIS IS BAD
*sshole is out
if you run sshd on alpine, openwrt or any musl box with password auth on: update now
official writeup: sshhole.com
Steering an agent swarm to rip through a multi-day plan on limited compute is hard mode -- the already saturated coordinator is forced to also coordinate compute allocation! I'm so sold on on-demand agent compute just to avoid this problem.
Tushar retweeted
Indians are being sold municipal failure as upward mobility: buying purifiers for unsafe water & toxic air, private transport as public transit fails. As incomes rise & capitalists institutionalise civic failures, accountability will decline & cities will deteriorate further.
Tushar retweeted
need a kalshi that bets on govt work timelines
#Traffic alert, #Bengaluru!
Traffic flow on Kempegowda International Airport (KIA) Road will be affected for 18 months from Nov 6 as BDA starts construction of a four-lane elevated corridor on Ballari Road. KSRTC/private buses and all goods vehicles are banned on the stretch.
Alternative routes 👇
@timesofindia
Yo @instamart_it you guys are losing real money to @letsblinkit by not letting people print on the web app @ instamart.in
Tushar retweeted
The Indian government is considering withdrawing the legal protection that currently shields social media platforms such as Facebook, Instagram, YouTube and X from liability for content posted by their users.
According to The Indian Express (November 5, 2026), the IT Ministry has begun discussions on changes to India's safe harbour framework under Section 79 of the Information Technology Act, 2000.
Currently, Section 79 protects online intermediaries from being held legally responsible for content uploaded by users, provided they meet certain legal requirements. This allows platforms to host millions of posts, videos and comments without automatically becoming legally responsible for each one.
The government is now exploring whether this protection should be limited to services that simply transmit information, such as internet service providers, telecom operators and peer-to-peer messaging platforms, rather than social media companies.
Why is the government considering this?
A section of the government believes social media platforms can no longer be treated as passive carriers of information. Platforms use algorithms to decide which posts appear in users' feeds, what content gets recommended and how widely it spreads. They also moderate content under their own community guidelines, which may not always align with Indian laws.
The government believes these practices raise questions about whether platforms should continue receiving the same legal protection as intermediaries that simply transmit information.
What developments led to this?
• In July, the government summoned Meta's global policy executives after a video uploaded by Prime Minister Narendra Modi was briefly withheld on Facebook in India during student protests.
• The government has raised concerns about deepfakes, content moderation and the reach given to certain political narratives.
• X has challenged the government's online content blocking and takedown framework, including the use of the Sahyog portal.
• The Supreme Court's October 15 observations in the Just Rights for Children Alliance vs S Harish case raised concerns about social media companies' failure to adequately prevent the circulation of child sexual exploitation and abuse material.
• A Parliamentary Standing Committee chaired by BJP MP Nishikant Dubey has also pushed for stronger action against social media companies and warned that Meta could lose its safe harbour protection.
What else is being considered?
The IT Ministry is also exploring two other changes:
• Requiring social media and messaging platforms to inform authorities before introducing major product changes that could affect user safety. This follows the government's intervention in WhatsApp's planned username feature.
• Strengthening grievance redressal mechanisms so users can seek action against deepfakes, sexually explicit content shared without consent and other forms of non-consensual imagery.
Has the government made a final decision?
No. Discussions are still at an exploratory stage. Removing or substantially changing Section 79 protection would likely require amendments to the Information Technology Act, 2000, which would need Parliament's approval.
Any major change to safe harbour could have significant consequences for how social media platforms operate in India and how online content is moderated.
