adi2907

What a transformation by Xiaomi to build almost frontier level models. Five years back, when I was in the data science team, they dint really bother about AI models and were using Baidu for NLP and vision under the hood of their APIs

This is definitely very useful, and something i've been thinking of to build myself. A personal CRM. Issue is when I see something like this - where it will take me around half hour to figure out - while i can vibe code something in an hour which will do much more and personalize it for me, I hold back on trying it out.

This is a 5 day - only code , no maths -introduction to data science covering all critical topics and suitable for beginners. As a practicing data scientist, this covers 80% of my toolkit. All datasets are created synthetically. I created this course as almost all the existing courses cover too much material, too much theory and maths and don't emphasize the practical aspect of using existing algorithms to achieve impactful output. I hope this helps beginners and I wish a course like this existed when I started years back to not dissuade me from learning this approachable field

Very impressed. I am not a scientist but am building a product for intent-based discounting in Shopify. Typically Google scholar gives me very generic results using LSTM etc however this search gave me some interesting results with focus on real world implementation. The clarifying questions are also quite impressive as it gives the impression that it is understanding the query really well. Good stuff. I think it might be useful for end-users and not just company/research folks as well

India's new data privacy law is quite simple to understand. Allows cross border data sharing with few countries. Stiff penalties for personal data breach or failure to report them

As an entrepreneur, I empathise with fellow startups losing customers due to mandatory check for recurring payments. However as a customer, this has been a godsend as I had almost 15 subscriptions totalling $300 monthly, quite a large amount in India.

The constant reminder of how many of those subscriptions are useless has allowed me to cut my expenses. Case in point, was subscribed to linkedin premium for last 2 years, while I make use of it only once in 3-4 months. Now I simply dont recharge my credit card and only do so once its required.

Not sure if its the ideal solution but definitely am thankful to it!

Wondering how can one build something massively successful which has GPT as the core engine. Will be ridiculously easy for anyone else to copy, if its trained on similar data as the use case

Surprised how this is in the front page. I had a similar feeling of bewilderment that I had when reading Das capital- fancy in theory but after several decades of wasting your/others life, discover it’s unimplementable

Here's a post on how he influenced companies halfway across the globe https://helloworld-adi.medium.com/delivering-happiness-an-ev...

Honestly haven't seen a clear answer till now. We had tried multiple ways to link high satisfaction scores with purchase behavior - correlation/regression of NPS with frequency/value of purchase.

At least in myntra case, the detractors turned out to be much more valuable than the entire universe, we put it down to customers who love us enough to hate us!

Really sad , I remember when I was setting up Myntra (Indian zappos) customer experience department, Tony and his book were a huge influence and everything that we did was measured up with the zappos way of customer experience. We were one of the few large companies having in house call centre with a five day week and training period of one month to all customer cate executives, plus a mandatory call listening session once a quarter for all senior leadership.

There are still some question marks on the ROI of trying to ensure customer delight especially via call center since discounts are such a huge factor in buying shoes and clothes online that even an NPS of 60 will not help if you don’t price lower especially for the value conscious Indian buyer however he definitely brought a new dimension of thinking for a lot of customer care folks

Have always marvelled at Amazon’s ability to do local Innovation at scale. Here is a behemoth for whom India is 10 percent of global revenue, yet launch a service which has nothing to do with their core e-commerce and cloud functions. Is it to get users, hiring tool or make more money. Will be interesting to learn the culture which facilities such decision making - any other large company would have shot down such a crazy idea long ago

Partially agree,,Google Colabs is outstanding, even though inspired by jupyter notebook. Taking over scientific community

China never allowed unfettered access to global firms. Since Deng Xiaoping opened up, the rule for a foreign firm to gain access was 1) to tie up with a local company which will have significant stake and 2) ensure technology transfer to the local company over a period. This was a pre condition for market access, and basically how electronics components makers were built. The recent stars like Huawei, Xiaomi and Lenovo could only be built once the supply ecosystem was local. You will not find a single example of any country becoming developed following free trade except for island nations like Singapore