Tushar Burman

Communicator. Ex JDM, D&B, Indian Express, Network18. Group Head, Adfactors PR. Youth icon.

[Tushar Burman @tburman]

Pune people, we’ve lost Genghiz. He’s a neutered stray, NOT fancy. If you know anyone who’s bought, please contact.

[Sudarshan Birla @birla_sb_]

Spent the last week building something I couldn't stop thinking about 😁. Mumbai has thousands of street cats (I love cats). They live in our lanes, outside our buildings, in most of the cafes.

So I built meowmbai - A community map where you spot, explore, & pin Mumbai's street cats.

Every cat gets a profile. Every neighborhood gets a count. People can upload cat sightings, discover cats around the city, explore adoption opportunities, & find helpful resources for cat care.

It's live. Go find your local cat. Have fun at
https://t.co/NbY1d7axLx 🐾🐈

#mumbaicats #meowmbai #cats #mumbai

[Tushar Burman @tburman]

@highspeedION your website and app do not accept my login. There's no way for me to pay.

[Tushar Burman @tburman]

@SriniKay Once it opens, I’m shifting to Delhi office. It’ll be quicker from NMIA.

[Tushar Burman @tburman]

The Atal Setu had a long honeymoon: about a year. Now I’m back to 2+ hour commutes between Lower Parel and Ulwe. There is no hope.

[Tushar Burman @tburman]

Passengers now unwilling to fly this plane. @flyspicejet 709. Apparently they’ve prevailed. We’re going back.

[Tushar Burman @tburman]

Two hours late and we’re still on the ground. Passengers revolting. @flyspicejet 709

[Tushar Burman @tburman]

@Uber_India bumper to bumper standstill traffic on Delisle bridge, but driver wants to stand on his horn. @MTPHereToHelp

[Tushar Burman @tburman]

@SriniKay I guess all the excess parking charges were worth it.

[Tushar Burman @tburman]

After years of using @getsimpl with a perfect track record and enhanced limits, they’ve summarily deactivated my account. No explanations, not even in a support response.

[Tushar Burman @tburman]

Great thread. Expecting omniscience from AI is a fundamental problem. Cite sources, or it didn’t happen. I like to think of my current tool workflow (Perplexity+NotebookLM+Claude) as a super-powered Ctrl/Cmd-F.

[Sridhar Vembu @svembu]

All the LLMs and other deep learning models are based on neural networks. We can think of them as mathematical functions with hundreds of billions of parameters.

Those parameters (weights) are determined during training and we train these networks with trillions of tokens (text, images, videos that are split up into tokens to be ingested by the models).

We can say that every one of the trillions of tokens played a part in determining the value of each of the hundreds of billions of parameters.

The image I have in mind is a giant lake where we dissolve trillions of cubes of salt, sugar etc. After the dissolution we cannot know which of the cubes of sugar went where in the lake - every cube of sugar is everywhere!

Therein lies a problem: if we use a business database, such as customer relationship data, to train a neural network model (i.e to determine its parameters), when the customer changes that data or deletes the data, we do not know how to alter the weights of the model to account for this change in the data. Even if the model were dedicated to that customer, we still cannot guarantee the customer that their changes to the data will be reflected in the model.

In that sense, neural networks (and therefore LLMs) are NOT a suitable database.

This is a fundamental limitation of the current scientific mathematical approach and cannot be fixed only by technological fine tuning.

The RAG (retrieval augmented generation) architecture keeps the business database separate and augments the user prompt with data fetched from the database.

In that case, the model itself is not trained on the (potentially changing) customer data because that data is only used in the prompt.

But RAGs can only go so far.

I personally have come to believe more foundational work is needed. What does that look like? All I have right now are hunches. That is the existing part of scientific work!

[Tushar Burman @tburman]

The future belongs to idea guys who can just do things https://t.co/j4ryqSKrWe #techblogs #feedly

[Tushar Burman @tburman]

https://t.co/yCr9DKEZw1

[Tushar Burman @tburman]

★ Don’t Throw the Baby Out With the Generative AI Bullshit Bathwater https://t.co/5wxFW7CeWL

[Tushar Burman @tburman]

https://t.co/mmstXUsWkz

[Rithwik Jayasimha @thel3l]

Apple released a hearing aids feature for the AirPods Pro a while ago. I bought a pair for grandma, but then realized that the feature was geoblocked in India

So we at @_lagrangepoint decided to unblock it. It ended up involving a leaky microwave and building a Faraday cage:

[Tushar Burman @tburman]

Journalism's fight for survival in a postliterate democracy https://t.co/JMT5O1bzwr

[Tushar Burman @tburman]

Question: what do you pay for electricity? I’m getting bills like ₹15000 from @MSEDCL with a single AC. 6000 without. I live alone. Entire building has this issue and officials say it’s because you use AC. Who pays half their rent in power bills?

[Tushar Burman @tburman]

25 minutes at elphinstone bridge signal. Things never change. @mumtraffic02

[Tushar Burman @tburman]

@pablochaterji As an editor, you likely write a lot less than you used to. But you’re able to get better results (hopefully) from your writers. You haven’t forgotten how to write, but you have gotten better at making others write better. The ats how I look at (and use) AI.

[Tushar Burman @tburman]

As a frustrated user having wiggled ram sticks until the pc boots for 30 years, I think I’m entitled to say that upgradability is not worth it beyond puberty. Solder everything. At least it has a better chance at a first and second life.