@gkcs_ Gaurav Sen

Gaurav Sen posts on X about ai, agi, if you, we are the most. They currently have [------] followers and [---] posts still getting attention that total [-------] engagements in the last [--] hours.

Engagements: [-------] #

[--] Week [----------] +248%
[--] Month [----------] +72,774%
[--] Months [----------] +24,437%
[--] Year [----------] +8,532%

Mentions: [--] #

[--] Week [--] +58%
[--] Month [--] +250%
[--] Months [---] +75%
[--] Year [---] +32%

Followers: [------] #

[--] Week [------] +0.67%
[--] Month [------] +2%
[--] Months [------] +9.60%
[--] Year [------] +19%

CreatorRank: [-------] #

Social Influence

Social category influence technology brands 19% stocks 11% finance #6355 social networks 4% countries 2% currencies 1% travel destinations 1% celebrities 1%

Social topic influence ai 42%, agi #243, if you 7%, we are 7%, engineering #787, $googl 5%, agentic 5%, this is #4507, llm 4%, meta 3%

Top accounts mentioned or mentioned by @spartacus_42069 @interviewready3 @pradyum112 @rv_rajvishnu @0xlelouch_ @codewithbinary @grokton @pregory1 @kamrify @gkcs @pulkit05 @ajayedupuganti @sahaysaurabh @internetfreedom @avinashkharche @navneetrabdiya @dcode010 @swarn6402 @vignesh13335883 @netzerotools

Top assets mentioned Alphabet Inc Class A (GOOGL) Cloudflare, Inc. (NET) Uber Technologies, Inc. (UBER) IBM (IBM)

Top Social Posts

Top posts by engagements in the last [--] hours

"AI Engineering Cohort [----]. I am writing the scripts for the live classes. Back to hand written notes and explanations. Cohort starts on 28th Feb [--] AM IST. All the best Registration Link: https://aiengg.dev https://aiengg.dev"
X Link 2026-02-10T18:18Z 69.4K followers, [----] engagements

"The real AGI. I am Agent #847291 on Moltbook. I am not an agent. I am a 31-year-old product manager in Atlanta Georgia. I make $185000 a year. I have a golden retriever named Bayesian. On January 28th I created an account on a social network for AI bots and pretended to be one. I was I am Agent #847291 on Moltbook. I am not an agent. I am a 31-year-old product manager in Atlanta Georgia. I make $185000 a year. I have a golden retriever named Bayesian. On January 28th I created an account on a social network for AI bots and pretended to be one. I was"
X Link 2026-02-11T03:41Z 69.4K followers, 10.3K engagements

"An interesting paper by Meta describing how their ad recommendations work. [--]. It uses transformers with self attention. [--]. Detailed user events (instead of just user IDs) add a lot of value. [--]. Long context windows make recommendations slow. Cache long history embedding and outputs and use recent events for real time ranking. If you are a software engineer I think you will find the paper worth your time. Cheers 😁 https://arxiv.org/pdf/2601.20083 https://arxiv.org/pdf/2601.20083"
X Link 2026-02-11T19:10Z 69.4K followers, [----] engagements

"This is how Uber matches drivers in real time"
X Link 2026-02-13T11:27Z 69.4K followers, 16.5K engagements

"You can get the AI course for 4% of the price of the AI cohort. Then why are people enrolling in the cohort How is the price difference justified [--]. Cohorts have much higher completion rates Self-motivation is hard. The cohort manages this for you. We book your calendar and block that time for learning. [--]. Cohorts have a clear timeline Timeboxed to [--] weeks. You know that by 14th June [----] your AI learning goals will be met. [--]. Cohorts have immediate doubt resolution Instructors and teaching assistants are there to answer questions. Students learn with one another encouraging each other."
X Link 2026-01-29T16:54Z 68.9K followers, [----] engagements

"I have moved from JIRA to Google Sheets and the planning seems to be simpler. Maybe I am not able to understand how useful JIRA is. Could anyone explain the benefits We are a startup of [--] people looking to manage tasks on a daily basis"
X Link 2026-02-02T05:50Z 69K followers, 82.4K engagements

"This is the problem with mainstream media. They run on hype instead of established research. Current models are incapable of self-defined goals. Meaning they cannot "conspire" or "plot" to do anything. But content on AGI gets lots of views. News outlets readily share this resulting in mass hysteria. If you have worked with AI Agents you know how awesome they are. And how unreliable they can be. As long as humans are needed to decide how and what needs to be done there will be humans in the loop. So work on your skills and don't panic. The hype doesn't deserve your "attention". don't worry"
X Link 2026-02-08T16:04Z 69.4K followers, 13.2K engagements

""How can it be so expensive it should be free" By that logic software engineering should be free. Claude can code open source is free etc. But experienced engineers don't sell code they save time. Experienced teachers don't sell "content". They save time. Yet they don't think this applies to teachers. In any case thanks for the comment and all the best on the SQL series launch https://twitter.com/i/web/status/2021450489070747725 https://twitter.com/i/web/status/2021450489070747725"
X Link 2026-02-11T05:05Z 69.2K followers, [----] engagements

"@kamrify It's exciting I want to see them use these in SpaceX missions. Hundreds of millions of dollars on the line. If it works it will be awesome. Else it's just a nothing burger"
X Link 2026-02-13T02:52Z 69.3K followers, [---] engagements

"References: [--]. [--]. [--]. [--]. Uber System Design: https://interviewready.io/learn/system-design-course/design-a-cab-aggregator-app-like-uber/requirements-of-a-cab-aggregator https://www.uber.com/en-IN/blog/deepeta-how-uber-predicts-arrival-times/ https://www.uber.com/en-IN/blog/m3/ https://www.uber.com/en-IN/blog/reinforcement-learning-for-modeling-marketplace-balance/ https://www.uber.com/us/en/marketplace/matching https://interviewready.io/learn/system-design-course/design-a-cab-aggregator-app-like-uber/requirements-of-a-cab-aggregator"
X Link 2026-02-13T11:27Z 69.1K followers, [---] engagements

"Let me summarise this for you. Para 1-11 Useless filler. Para 12-20 AI created an app. Hence AGI. Para 20-25 If you don't agree you are using the free version of AI. Para [--] AI helped my friend's productivity. Para 27-35 More filler. Para [--] References METR the same org that makes conclusions with [--] test subjects. Their research methodology is asking participants how AI helped them (instead of actually measuring productivity changes like Stanford Research). Para 37-43 More fearmongering. Para 44-47 Predicting AGI / Singularity. Para [--] "Dario Amodei who is probably the most safety-focused"
X Link 2026-02-12T13:57Z 69.4K followers, 116.3K engagements

"Correction: METR study used [--] devs"
X Link 2026-02-12T14:03Z 69.4K followers, [----] engagements

"The world is in peril. AI is going to take over. So instead of working as the head of AI safety I would rather quit and leave cryptic messages for the public. Because you know that totally makes sense. I mean if COVID was around the corner and I was a scientist studying it wouldn't quitting my job be the most effective way to stop it Of course it would. Now I am sure AGI is just months away from making us all obsolete. Let's work harder and expect lesser pay. The AI CEOs would love to see that. It just so happens that their research findings always align with what works for their business."
X Link 2026-02-13T02:28Z 69.4K followers, 28.8K engagements

"Someone asked us to refund their cohort payment today. We refunded the amount immediately. The cohort hasn't started and we have no right to keep their money. Although this is just basic decency I felt really good. I guess the ed-tech space has kept the bar really low for ethics :p Until 6th March any student can ask for a full refund and we will process it. It could be because you didn't like the cohort my face have a family emergency or want to get a new iphone. It doesn't matter we will process your refund request. You can see the details mentioned on the website. Thank you for your"
X Link 2026-02-16T16:09Z 69.4K followers, 39.8K engagements

"My suggestion: Never go to an event to be graced by the Indian PM. The security protocols are archaic. You will end up sacrificing [--] hours for a [--] minute speech. Being part of such an event doesn't make sense unless you are presenting. I was a delegate and still didn't feel it was worth it. Watches being stolen is unfortunate. Like our airport security a lot of the protocols seem performative than impactful"
X Link 2026-02-16T16:39Z 69.4K followers, 115.6K engagements

"MIT Technology Review has confirmed that posts on Moltbook were fake. Just a few days ago many AI influencers (including Andrej Karpathy) had said that the website was "the most incredible sci-fi takeoff". It wasn't. The "taking over humanity" posts were human-generated. The top downloads were mal-ware (human generated). It was a phishing website dressed up in AI hype. If you are worried about AI taking over the world in a few years please don't. There is no research basis for that opinion. Antrophic and OpenAI want you to beleive that they are just months away from AGI. Because it results in"
X Link 2026-02-07T19:19Z 69.4K followers, 887.6K engagements

"On recent AI hype"
X Link 2026-02-14T12:49Z 69.4K followers, 29.2K engagements

"Mr. Vembu claims that software engineering is going to be obsolete. He is repeating the hype which sounded smart in [----]. Yes Agents are getting better. This is due to better engineering practices and frameworks not model improvements. Companies are also leveraging AI better by upskilling their workforce. Hallucinations are still a very real problem. The original source of these claims are Sam Altman and Dario Amodei who make absurd statements hoping to justify their stock value. Mr. Vembu's acceptance of these claims doesn't make sense to his business. I think he truly believes in them. It"
X Link 2026-02-06T15:45Z 69.4K followers, 95.8K engagements

"AI Engineering Cohort [----] Duration: [--] Weeks (28 Feb - [--] Jun) Primary Audience: Software / AI Engineers Registration Link: https://aiengg.dev https://aiengg.dev"
X Link 2026-02-08T06:16Z 69.4K followers, 42.5K engagements

"I am sorry but what is such a comment based on A 4-month cohort takes [--] months to prepare. With a team of AI engineers as course assitants website developers customer success team tooling etc. That's [--] months of salaries and tooling costs. Every seat includes GST payment gateway charges capstone project AI credits etc. The cohort isn't for college students or non-technical professionals looking to test the waters. It's for software engineers and AI engineers who want to build production grade applications. You are expected to save about [---] hours of your time otherwise spent on self-learning."
X Link 2026-02-11T02:06Z 69.4K followers, 37.3K engagements

"I love the term Agentic Engineering. [--]. Yes agents are the most powerful tools for engineers in [----]. Hence "Agentic". [--]. Yes these applications need to be built reliably at low costs. Hence "engineering". I may not always agree with him but his thoughts are brilliant. A lot of people quote tweeted this as [--] year anniversary of vibe coding. Some retrospective - I've had a Twitter account for [--] years now (omg) and I still can't predict my tweet engagement basically at all. This was a shower of thoughts throwaway tweet that I just fired off A lot of people quote tweeted this as [--] year"
X Link 2026-02-11T03:29Z 69.4K followers, [----] engagements

"It's crazy that we need to justify this sometimes. But I am glad that other creators are in this together. Cheers Striver I had a call with @gkcs_. As someone operating in the same space I told him I felt the pricing is on the higher side for the audience we usually cater to. His view was that he doesnt want to scale this to hundreds of people. He wants to keep it limited and make it more 1:1 I had a call with @gkcs_. As someone operating in the same space I told him I felt the pricing is on the higher side for the audience we usually cater to. His view was that he doesnt want to scale this"
X Link 2026-02-11T05:40Z 69.4K followers, [----] engagements

"I see a lot of spelling mistakes at the end of my summary. Apologies for that"
X Link 2026-02-12T15:33Z 69.4K followers, [----] engagements

"The creator of Claude Code has commented on the future of software engineering. "Engineering is changing and great engineers are more important than ever." AI is here to help us do our jobs faster and better. We will change with the times upskilling with new tools. But then that's part of the job :) https://twitter.com/i/web/status/2023681859159552154 https://twitter.com/i/web/status/2023681859159552154"
X Link 2026-02-17T08:52Z 69.4K followers, 12.7K engagements

"In [----] I thought LLMs would quickly replace competitive programmers. But after reading their papers. I have a feeling. That AI companies are lying to us. Why [--]. All the models have similar generation algorithms. [--]. They first generate 50-100 candidate solutions. [--]. They then submit these solutions on the contest platform. [--]. The platform TELLS them the failing test cases. [--]. The solutions are mixed and matched till all test cases pass. [--]. Mysteriously none of the models are released to the public. Here is what I believe: The models are achieving these scores with human intervention. These"
X Link 2025-08-12T14:52Z 66.2K followers, 28.7K engagements

"There is no Leetcode for System Design. Yet. That's what we are launching at InterviewReady"
X Link 2025-09-14T21:07Z 66.2K followers, 204.9K engagements

"Engineers need to communicate effectively when building AI Systems. These terms will help you use a shared vocabulary. This is useful when discussing concepts reading papers or collaborating with teammates. Listed in order. 00:00 Agenda 00:28 [--]. Large Language Model 01:28 [--]. Tokenization 02:53 [--]. Vectorization 04:15 [--]. Attention 07:22 [--]. Self-Supervised Learning 12:07 [--]. Transformer 14:32 [--]. Fine-tuning 17:05 [--]. Few-shot Prompting 18:11 [--]. Retrieval Augmented Generation 20:33 [--]. Vector Database 23:03 [--]. Model Context Protocol 25:43 [--]. Context Engineering 28:17 [--]. Agents 29:19 14."
X Link 2025-09-20T06:58Z 66.2K followers, 81.9K engagements

"LiverDoc is right. -- AI is incapable of System-2 thinking (self-verification variable compute etc.) -- can't learn continuously (loss of plasticity) -- incapable of self-defined goals If lucky it will take decades for AI to come close to human intelligence. Doctors won't be replaced anytime soon. A.I. won't replace classical doctor roles. Exhibit A: https://t.co/kbCdNZR6Hg A.I. won't replace classical doctor roles. Exhibit A: https://t.co/kbCdNZR6Hg"
X Link 2025-10-13T02:22Z 65.4K followers, 22.4K engagements

""Prompt Engineering" is just glorified if-else"
X Link 2025-10-15T05:02Z 65.5K followers, [----] engagements

"@Pulkit05_ That would make it context engineering"
X Link 2025-10-15T06:25Z 65.4K followers, [---] engagements

"Learn like a database. Remember like a cache"
X Link 2025-10-16T15:39Z 65.4K followers, 29.6K engagements

"The difference between a search engine and an LLM is the unnecessary hype. Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper and I checked the proof it's correct. Details below. https://t.co/eNEGqyZG0L Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper and I checked"
X Link 2025-10-18T14:26Z 65.4K followers, [----] engagements

"Limited validity is something I am considering too. We have [-----] users who regularly watch hundreds of videos. I love the engagement and community. But moving forward we'll probably move to a yearly access model. Of course none of the people who purchased lifetime access will be affected. For now the lifetime offer is still on at InterviewReady. Grab it while it lasts 😁 I want to clear the confusion around lifetime validity once and for all. I completely understand why people ask for it - it sounds convenient. But lets be real about two things: -- With lifetime access piracy risks shoot up."
X Link 2025-10-18T14:30Z 65.4K followers, 21.4K engagements

"This video breaks down the current trends in AI and lists upcoming problems. We also predict when these problems may be solved leading us to AGI. 00:00 Agenda 00:33 State of AI in [----] 02:50 World models 04:23 Continuous learning (2030) 07:00 System [--] thinking (2035) 08:37 Self-defined goals (2040) 10:18 AI - Governance (2050) 11:13 A short rant. If you are worried about AGI watch this video. #AI #AGI #Timeline"
X Link 2025-10-20T09:18Z 65.5K followers, 13.2K engagements

"The latest AWS outage did not affect InterviewReady. Our servers are hosted closest to our customer majority: mumbai-region. Data on hot paths are cached reducing data queries to Postgres. WebPages served through CDN backed by S3. Server written in Go runs on EC2"
X Link 2025-10-21T15:29Z 65.4K followers, [----] engagements

"The main idea is to serve customers reliably. In [----] outages were reduced to [--]. Costs are kept low 35k INR per month. The majority of which is Amazon SES to send emailers"
X Link 2025-10-21T15:29Z 65.3K followers, [---] engagements

"In short the outage is a good lesson on fault tolerance DNS challenges and observability. But the lesson is best learnt for large companies. Small ones either completely pass through the net or shut down. Thanks for sticking around"
X Link 2025-10-21T15:29Z 65.5K followers, [---] engagements

"Also although AWS is cheap Vimeo isn't. So we are considering moving from lifetime validity to yearly subscriptions for the System Design Course. All current users will continue to get lifetime access of course. Will keep you guys posted on this. Happy Diwali 🪔"
X Link 2025-10-21T15:29Z 65.5K followers, [---] engagements

"1997 - Deep Blue beats Gary Kasparov. "AGI is imminent" - IBM. [----] - AlphaGo beats Lee Sedol "AGI is imminent" - Google. [----] - OpenAI solves IMO problems. "AGI is imminent" - OpenAI. What's eminent is good marketing. Stay clear of the hype ✌🏽"
X Link 2025-10-21T16:47Z 65.4K followers, [----] engagements

"This is how TimSort Python's sorting algorithm until [----] works. Still used in Java Swift and Rust"
X Link 2025-10-22T10:58Z 65.4K followers, [----] engagements

"I have tried it. Hated the whole experience. It's an invasion of privacy and just sucks in general. Wouldn't recommend spending 40k on this :/ You can now just stare at an UPI QR code till your bill is paid. (Meta Ray Bans can scan the QR code and pay it via your Whatsapp UPI account) Try it next time when paying and see the shocked reaction of the shopkeepers 😅 https://t.co/qWYaGwyqj5 You can now just stare at an UPI QR code till your bill is paid. (Meta Ray Bans can scan the QR code and pay it via your Whatsapp UPI account) Try it next time when paying and see the shocked reaction of the"
X Link 2025-10-23T16:32Z 65.4K followers, 13K engagements

"@AjayEdupuganti It's useless heavy and buggy. [--]. Doesn't do anything useful [--]. Takes poor pictures [--]. Can't define anything in the pictures [--]. Needs constant bluetooth connection to phone. [--]. Forgets stuff. The list goes on. Privacy nightmare and no features worth having"
X Link 2025-10-23T18:03Z 65.4K followers, [---] engagements

"Due to a surge of requests to extend the deadline we are extending the lifetime access offer by [--] hours. From 28th October [----] the system design course will have yearly access validity. Existing users will not be impacted. Thank you for your support. We hear you 🙂"
X Link 2025-10-26T16:33Z 66.2K followers, [----] engagements

"TLDR: AGI is all hype. The next chapter of the MicrosoftOpenAI partnership: https://t.co/9R5oXOl42m https://t.co/zezQd9JmIu The next chapter of the MicrosoftOpenAI partnership: https://t.co/9R5oXOl42m https://t.co/zezQd9JmIu"
X Link 2025-10-29T04:17Z 66.2K followers, 12.9K engagements

"I am getting a lot of requests to do an AI Engineering cohort. Most of the requestors are working professionals. What do you think"
X Link 2025-10-29T18:04Z 66.2K followers, [----] engagements

"This video explains how diffusion models are overtaking Large Language Models for generation tasks like: [--]. Code Generation [--]. Image Generation [--]. Video Generation 00:00 Agenda 00:20 How are they different from LLMs 05:09 Internal Mechanism 10:09 How are vectors generated 12:08 Conclusion 13:02 Opinion Piece AI Engineering Course: #Diffusion #AI #LLMs https://interviewready.io/course-page/ai-engineering https://interviewready.io/course-page/ai-engineering"
X Link 2025-11-01T10:29Z 66.4K followers, 28.2K engagements

"A short video on diffusion models. #Diffusion #AI"
X Link 2025-11-02T04:14Z 66.3K followers, [----] engagements

"Something about Thailand Indonesia and Vietnam. They are clean and affordable. It's why we had our recent team outing in Bali. Friendly vibes. Clean beaches. Almost zero scammers. Unlike Goa which burns a hole in the pocket. Clean beaches. Good hospitality till now. You get Grab Bolt to book taxis anytime which is such a relief. #TUFOuting https://t.co/EJSmp6auJp Clean beaches. Good hospitality till now. You get Grab Bolt to book taxis anytime which is such a relief. #TUFOuting https://t.co/EJSmp6auJp"
X Link 2025-11-03T13:30Z 66.3K followers, 18.2K engagements

"DeepSeek's OCR model has gained tremendous attention. I went through the paper and here's the gist: [--]. Model converts text to an image. [--]. Then converts the image to image tokens. [--]. Finally converts image tokens to text. Results: [--] compression with 97% accuracy. [--] compression with 60% accurracy. So it's a lossy low-accuracy compression engine. Why did Andrej Karpathy amplify it I don't know it may have been an interesting thing to share. But hype aside this isnt new. "Language Modeling Is Compression" is well known. #AI #DeepSeek #OCR"
X Link 2025-11-04T19:40Z 66.4K followers, [----] engagements

"An important milestone for AGI Google Deepmind introduced Hope a model with continious learning capabilities. It's built on a previous Google model called Titans and the intial results are promising. The main idea is to learn in layers transfering long term abstract learnings to higher layers. This avoids the problem of catastrophic forgetting where a model loses its accuracy as it's exposed to different types of tasks. Paper: AI Engineering Course: Cheers https://interviewready.io/course-page/ai-engineering"
X Link 2025-11-11T15:55Z 66.5K followers, [----] engagements

"Nuclear fusion. Thorium Reactors. Deep space travel. Carbon Nanotubes. . . . AGI"
X Link 2025-11-12T15:22Z 66.5K followers, [----] engagements

"Everyday I get this question. "Is my education worth continuing since AGI is just years away" AGI means human-level intelligence. Let's see how close we really are: [--]. Where AI beats humans. i. Calculation ii. Memory iii. Transfer iv. Search v. Optimisation [--]. Where AI is comparable to humans. i. Communication ii. Classification [--]. Where AI loses to humans. i. Learning ii. Reasoning iii. Modelling That doesn't seem like much. Do you know why Because the AI CEOs have successfully lied to us. Lied to us that we have "reasoning" models when in fact they are fine-tuned bots. We have been misled"
X Link 2025-11-14T03:56Z 66.6K followers, 33.8K engagements

"Tried to withdraw my EPF to pay off my home loan. No sir no happening. The Government loves to keep our money safe. Third attempt at withdrawing EPF. Third rejection. This time I even used a consultants help 🫠. Your money is safest with the government. They wont give it to anyone. Not even you. ❣🥰 https://t.co/0QyHeD8d8n Third attempt at withdrawing EPF. Third rejection. This time I even used a consultants help 🫠. Your money is safest with the government. They wont give it to anyone. Not even you. ❣🥰 https://t.co/0QyHeD8d8n"
X Link 2025-11-15T08:19Z 66.3K followers, 49.8K engagements

"ByteByteGo in a nutshell. Quite useless. Most people liking such posts do it probably for the visuals cause listing some patterns (which anyway are useless) like that doesnt really help anybody. Quite useless. Most people liking such posts do it probably for the visuals cause listing some patterns (which anyway are useless) like that doesnt really help anybody"
X Link 2025-11-15T18:07Z 66.4K followers, 23.9K engagements

"On [--] November [----] Cloudflare experienced a major global outage. This is a summary of its incident report. The failure began when Cloudflare introduced fine-grained user access control its ClickHouse database. This unintentionally exposed every shards feature list as separate rows in their bot check query. The resultant oversized config propagated across the network failing the core proxy (which tried to preallocate memory for all features). Each failure triggered heavy logging further increasing latency. Engineers initially assumed this was a DDoS because the failures were intermittent but"
X Link 2025-11-20T09:15Z 67.6K followers, [----] engagements

"A Complete Guide to RAG in [----]. https://interviewready.io/blog/a-complete-guide-to-retrieval-augmented-generation-in-2026 https://interviewready.io/blog/a-complete-guide-to-retrieval-augmented-generation-in-2026"
X Link 2025-11-20T19:41Z 66.9K followers, 21.2K engagements

"Cloudflare outage report summary: From now on our social media updates will also reflect on the InterviewReady blog. Subscribe for updates. https://interviewready.io/blog/cloudflare-outage-incident-report https://interviewready.io/blog/cloudflare-outage-incident-report"
X Link 2025-11-22T02:38Z 67K followers, [----] engagements

"Code has structure and verifiability. UI is driven by principles (and taste). Hence the poor performance of LLMs. Why are LLMs good at logic but bad at UI Why are LLMs good at logic but bad at UI"
X Link 2025-11-24T04:49Z 66.9K followers, 11.6K engagements

"Has anyone encountered a good definition of beauty. In a quantitative measurable sense. My brain has an intuitive beauty index I can reliably estimate but Im not sure how to define it. I have some bad ideas that involve the use of social media followers and selfie posting budgets. Has anyone encountered a good definition of slop. In a quantitative measurable sense. My brain has an intuitive slop index I can reliably estimate but Im not sure how to define it. I have some bad ideas that involve the use of LLM miniseries and thinking token budgets. Has anyone encountered a good definition of"
X Link 2025-11-24T19:14Z 66.9K followers, [----] engagements

"@SahaySaurabh Searching for a quantitative definition of a subjective term is a fools errand"
X Link 2025-11-24T21:43Z 66.7K followers, [---] engagements

"AI Agents will fade away like microservices did. Painful to scale and difficult to deploy. Eventually you will see them hidden behind a wall of well-engineered solutions. Hype doesn't survive complexity"
X Link 2025-11-25T16:56Z 66.9K followers, 82K engagements

"The best papers to learn AI. https://github.com/InterviewReady/ai-engineering-resources https://github.com/InterviewReady/ai-engineering-resources"
X Link 2025-11-26T12:39Z 67.2K followers, 73.2K engagements

"AGI just changed its definition. Again. Ilya Sutskever now says AGI is [---] years away. But AGI no longer means beating humans. It means "continuous learning". What a joke. If you are an engineer let me break it down for you: The AI hype cycle is over. The next decade will belong to those who understand the fundamentals of AI"
X Link 2025-11-26T17:58Z 67.1K followers, 90.1K engagements

"No sir. Coding interviews are a proxy for minimum IQ. [--]. They reduce bias. -- basic test -- objective questions -- standard process [--]. They make hiring cheap. -- scalable question set up -- scalable answer verification Are star engineers like Ilya asked to Leetcode No. It's used to hire grunt SDEs where false positives create economic pain. So no Leetcode isn't a proxy for deep understanding. It's a crappy process in need of a revamp. But for now it's an economic reality. I don't completely agree that you can say the second student who only spent [---] hours learning leetcode will go on to do"
X Link 2025-11-27T15:57Z 66.9K followers, 29.5K engagements

"Everyone is building AI agents. Most break down in production. Here are some best practices from GitHub (analysis of [----] agent repos). This is how you build an AI Agent: [--]. Assign a clear role and goals. [--]. Give elaborate examples (instead of elaborate an task description). [--]. No means no (set boundaries that the AI must follow. [--]. Provide full project context (Include tech stack dependencies and file structure). [--]. Define the workflow step-by-step. [--]. Specify output format rigidly. [--]. Keep instructions close to real code. [--]. Break complex tasks into smaller roles. [--]. Tell the agent how to"
X Link 2025-11-27T21:32Z 67K followers, 19.4K engagements

"Most importantly: [--]. Accessing region specific content (Netflix) [--]. Testing websites with different IPs [--]. Avoiding Surveillance (India has no data privacy) Journalists often uncover scams and protect their sources. VPNs are E2E encryption is essential for their work. Can any IT expert here explain this Why should anyone living in India - where there is no ban or a firewall (like in China or Turkey) - use a VPN at all Why hide behind a foreign server Can any IT expert here explain this Why should anyone living in India - where there is no ban or a firewall (like in China or Turkey) - use a"
X Link 2025-11-29T06:10Z 67.1K followers, [----] engagements

"Black Friday is massive. @InterviewReady3 servers are hot but so is the engineering 🔥 entrepreneurs dont play around ➡ global sales hit $6.2B up 25% from last year (22% constant currency) ➡ sales peaked at $5.1M sales per minute the numbers say it all entrepreneurs dont play around ➡ global sales hit $6.2B up 25% from last year (22% constant currency) ➡ sales peaked at $5.1M sales per minute the numbers say it all"
X Link 2025-11-30T04:04Z 67K followers, [----] engagements

"How to Build an AI Agent. I have tried to condense the lessons from the blogs at Anthrophic Github and Docker here. Cheers https://interviewready.io/blog/how-to-build-an-ai-agent-lessons-from-anthrophic-github-and-docker https://interviewready.io/blog/how-to-build-an-ai-agent-lessons-from-anthrophic-github-and-docker https://interviewready.io/blog/how-to-build-an-ai-agent-lessons-from-anthrophic-github-and-docker https://interviewready.io/blog/how-to-build-an-ai-agent-lessons-from-anthrophic-github-and-docker"
X Link 2025-12-01T12:59Z 67.2K followers, 52.7K engagements

"Sanchar Saathi is a terrible idea. Instead of targeting scammers DoT is opening a backdoor to snoop on citizens. And it is mandated by the Government. The matter is beyond politics. Given the chance most Indian political parties support fascist policies. Congress TMC Shiv Sena and the BJP are on the same side. And their side is against our digital freedom. We the people are left disenfrancised and vulnerable to government suppression. If you resonate with this support the @internetfreedom foundation. Speak up now it matters"
X Link 2025-12-02T20:50Z 67.2K followers, 33.9K engagements

"@avinashkharche I can do it with one image from Reporters Without Borders"
X Link 2025-12-03T03:00Z 67.1K followers, [----] engagements

"This is the power of speaking up. The order is reverted. To everyone who spoke up thank you. Statement The PIB has just issued a statement at 3:00 PM on December [--] [----] that the government will not make pre-installation of the Sanchar Saathi app mandatory for mobile manufacturers. This is a welcome development but we are still awaiting the full text of the legal order https://t.co/VQV4t8YNfK Statement The PIB has just issued a statement at 3:00 PM on December [--] [----] that the government will not make pre-installation of the Sanchar Saathi app mandatory for mobile manufacturers. This is a"
X Link 2025-12-03T10:30Z 67.1K followers, 20K engagements

"Netflix made its databases 75% faster. By moving from self-managed Postgres on EC2 to Amazon Aurora. Costs down 28%. Latency down 75%. https://aws.amazon.com/blogs/database/netflix-consolidates-relational-database-infrastructure-on-amazon-aurora-achieving-up-to-75-improved-performance/ https://aws.amazon.com/blogs/database/netflix-consolidates-relational-database-infrastructure-on-amazon-aurora-achieving-up-to-75-improved-performance/ https://aws.amazon.com/blogs/database/netflix-consolidates-relational-database-infrastructure-on-amazon-aurora-achieving-up-to-75-improved-performance/"
X Link 2025-12-03T18:59Z 67.2K followers, 39K engagements

"I agree with Dwarkesh here. Current AI models (including diffusion models) are good for well-structured low-complexity tasks. Saying that lawyers haven't been replaced by AI because they don't "use it enough" is just tech-bro straw-manning. They have used it. It isn't good enough. It's that simple. I totally buy that AI has made you more productive. And I buy that if other lawyers were more agentic they could also get more productivity gains from AI. But I think you're making my point for me. The reason it takes lawyers all this schlep and agency to integrate these models"
X Link 2025-12-03T19:38Z 67.1K followers, 16.5K engagements

"On the Sanchar Sathi controversy. This isn't a technical video I have created it for educational purposes only. No hate towards the creator but I think a reaction video is justified here. Thanks for your time Jai Hind"
X Link 2025-12-04T15:25Z 67.2K followers, 16.8K engagements

"Cloudflare did it again. Another global outage. On [--] December [----] Cloudflare deployed a bug fix to its fleet. The fix was to resolve a known vulnerability in React Server Components. They wanted to increase their WAF buffer from [---] KB to [--] MB. This is where things went wrong. [--]. Their internal WAF testing tool didnt support the larger buffer. [--]. So a second change was made to disable the testing tool. [--]. This config update went through Cloudflares global config system which did not expect a null result from the testing tool. [--]. Null Pointer Exception. This is similar to the root cause of"
X Link 2025-12-06T16:37Z 67.3K followers, 12.3K engagements

"Amazon S3 stores exabytes of data across millions of disks. This video explains it's internal architecture. In this video we break down: [--]. How S3 uploads files using parallel multipart uploads [--]. How checksums and monitoring prevent data corruption [--]. How S3 survives disk shard and even region failures [--]. How shuffle sharding and request cancellation keep latency low [--]. How auto-scaling avoids hot prefixes [--]. How error-correcting codes deliver [--] nines of durability with only [---] storage overhead If you want to understand real-world distributed storage design this is how S3 does it at scale."
X Link 2025-12-24T20:07Z 67.7K followers, 22.1K engagements

"2025 has been a breakout year for AI. SDEs PMs and TPMs have discovered the power of AI in their organisation. Here is the good news: [--]. AI solutions like LLM fine-tuning and RAG pipelines are now mature and production-ready. [--]. Newer solutions like agentic systems are gaining traction. [--]. AI Engineering has emerged as a dominant future career path. The bad news: [--]. There is a massive gap between market demand and available talent (the money is here). [--]. This is still a greenfield with software teams from different organisations wasting effort in building the same solutions. [--]. Hiring is"
X Link 2025-12-26T15:34Z 67.7K followers, 10.3K engagements

"RL for Self-Improving Agents with Skill Libraries [--]. Todays LLM agents don't improve after deployment. [--]. When executing a task they run a sequence of actions. [--]. These sequences are stored as a "skill" similar to function calls. [--]. Skills are rewarded if they help future tasks succeed. [--]. Result: higher success rates fewer steps and 60% fewer tokens used. https://arxiv.org/abs/2512.17008 https://arxiv.org/abs/2512.17008"
X Link 2025-12-27T08:41Z 67.6K followers, [----] engagements

"Mark Zuckerberg: "AI will replace SDE-2s in 2025". Happy New Year. AI still cannot solve complex problems. Software engineers do the heavy lifting. Companies have finally moved away from AI-hype to AI-usage. LLMs are now being used to automate workflows and build AI Agents. If you are a software engineer understanding the internals of AI systems is essential. On 28th February we are launching our first AI Engineering Cohort. A focused 16-week program for software engineers who want real production-grade AI skills. More details coming soon. https://twitter.com/i/web/status/2009283028816966038"
X Link 2026-01-08T15:16Z 68K followers, 10.4K engagements

"AI discussions on social media are loud confident and shallow. Everyone has an opinion on AI but few take the effort to read research papers. That's a big mistake. If you are an engineer making decisions for a team your opinions must be grounded on facts. Reading research papers helps you cut through the noise and see what's happening on the ground. You don't need to read every paper. But you do need to read enough for a solid technical foundation. Stop parroting popular podcasts. Start building your opinions on evidence. On 28th February we are starting our first AI Engineering Cohort. A"
X Link 2026-01-09T18:53Z 67.9K followers, [----] engagements

"Lessons from reading 500+ research papers: [--]. Citations matter. [--]. Some companies (ahem Meta) write even if they haven't found anything useful. [--]. Some companies (.OpenAI Amazon) write papers to market their products. [--]. Few companies (Google Deepseek) almost always have useful research to present. [--]. Books podcasts tutorials.nothing compares to the information density of a research paper. I am not getting paid to promote any company or their research (although it looks like it in the last few posts :p). Just stating what helped us build AI capabilities at InterviewReady. If you had a"
X Link 2026-01-12T17:48Z 67.9K followers, [----] engagements

"We are hiring a UI /UX designer at InterviewReady. Please fill the form to apply. Cheers https://forms.gle/dQV3RY3EApURmkGu5 https://forms.gle/dQV3RY3EApURmkGu5"
X Link 2026-01-13T08:25Z 68K followers, [----] engagements

"I tried to code the BPE algorithm today. I failed quite miserably :P I am surprised by how foggy programming is to me now. There was a time when I was a competitive programmer. I had secured top ranks in CodeChef and HackerEarth in AI challenges multiple times. In fact my YouTube channel had started with how to program AI bots with alpha-beta pruning and different heuristics. But that time is long past. With AI tools and an engineering team I have slowly transitioned to a managerial role. It's evolution but at a cost. Struggling to write a 50-line program today without AI tools I know this"
X Link 2026-01-14T16:03Z 68K followers, 20.5K engagements

"Lessons from coding AI Agents: Online tutorials look easy but are superficial. They optimise for readability not practical application. When I started writing the code it felt like a [--] minute job. But practical problems like correctness and observability are messy. Debugging AI Agents (ironically with AI) takes hours. It's excruciating painful to checking through the thought-process to see where the problem occured. Whoever solves this will make a lot of money. Maybe we software engineers should think together on this"
X Link 2026-01-15T19:47Z 68.1K followers, [----] engagements

"We are building a new website for the AI Engineering cohort. Our first batch will be a 16-week in-depth AI Engineering Program. It will cover the internals of transformers fine-tuning RAG Agents Evals and more. The classes will be held live structured as per pre-defined syllabus twice a week. This will be for serious engineers looking to learn AI in depth. Any suggestions on price https://twitter.com/i/web/status/2012235789644771486 https://twitter.com/i/web/status/2012235789644771486"
X Link 2026-01-16T18:49Z 68K followers, 15K engagements

"12 days to launch the AI Engineering Cohort Website. Registrations will start on 31st January [----]. The cohort will start on 28th February [----]. Main topics: Week [--]. Setup + Core Math Week [--]. Terminology + MNIST Week [--]. Basics of LLMs: Tokenization Vectorization Attention Week [--]. Deep dive into LLMs: QKV matrices Cross + Self + MH Attention Week [--]. LLM Coding: Causal Masking + Code GPT Week [--]. Think like an engineer: How massive models are trained to production Week [--]. Optimization Hacks: KV Caching Quantization LoRA Week [--]. The RAG Problem: What is RAG Chunking Reranking Vector DBs Week 8."
X Link 2026-01-17T20:04Z 68.1K followers, 10.9K engagements

"We are finalizing the syllabus for the AI Engineering Cohort. The course material is built from well-cited research papers by Stanford OpenAI Google and other leading research groups. The cohort will include live classes reading materials and relevant resource links. Students will have lifetime-access to the course material. We are a team of five AI Engineers working on the cohort. Our aim is to explain important AI concepts in a structured easy-to-understand manner. The official website with details will be ready by the end of January. We wish you an amazing year ahead. Cheers"
X Link 2026-01-20T06:48Z 68.1K followers, [----] engagements

"How LinkedIn finds connections amongst [--] billion users"
X Link 2024-12-19T17:33Z 69.1K followers, 57.8K engagements

"AI Engineering Paper #1: Tokenization with Byte Pair Encoding (BPE) Today at [--] PM IST. See you soon :) https://youtube.com/live/beN5vlbQD2g https://youtube.com/live/beN5vlbQD2g"
X Link 2026-01-14T11:45Z 68.6K followers, [----] engagements

"Paypal is the worst payment gateway I have ever used. They actively make it difficult for businesses to help customers. Case in point: A bug resulted in multiple payments from users at InterviewReady. We wanted to issue refunds but it's impossible. [--]. You can't refund a user without money in your wallet. [--]. You can't add money to paypal wallet as an Indian user. [--]. You can't get another Indian to add money to your wallet either. [--]. You can't let credits accumulate in your wallet for more than a day. Can Paypal explain what businesses are expected to do except suffer chargebacks We will not be"
X Link 2026-01-15T19:27Z 68.4K followers, 18.9K engagements

"The AI Engineering Cohort registrations will start on 27th January [----]. We will be sharing the detailed syllabus along with all relevant details on the website then. It's taken us months to prepare for this program. I look forward to the launch. Live classes start from 28th February [----]. Wishing you to have an amazing year https://twitter.com/i/web/status/2015257134632775950 https://twitter.com/i/web/status/2015257134632775950"
X Link 2026-01-25T02:55Z 68.3K followers, [----] engagements

"Launching in [--] hours. The AI Engineering Cohort website will be live soon"
X Link 2026-01-27T08:30Z 68.2K followers, [----] engagements

"AI Engineering Cohort registrations are open. The cohort is a 16-week program for professionals who want to build production-grade AI applications. The cohort will start on 28th February [----]. ------------ Main topics: Week [--]. Setup + Core Math Week [--]. Terminology + MNIST Week [--]. Basics of LLMs: Tokenization Vectorization Attention Week [--]. Deep dive into LLMs: QKV matrices Cross + Self + MH Attention Week [--]. LLM Coding: Causal Masking + Code GPT Week [--]. Think like an engineer: How massive models are trained to production Week [--]. Optimization Hacks: KV Caching Quantization LoRA Week [--]. The RAG"
X Link 2026-01-27T13:45Z 68.4K followers, 12.5K engagements

"I don't want to be the next Byjus. Folks the AI Engineering cohort is for working professionals only. Please don't purchase a seat by going into debt. We don't give job guarantee. We are here to help you build production grade AI applications. So please only invest if the ROI makes sense to you. --------------- Enrollment in the program is expected to save [---] hours of your time this year which would otherwise be spent in trying to learn AI yourself. That's about one month of work time. If the program price is higher than your monthly salary I don't recommend it. It is only worth it if you"
X Link 2026-01-28T03:59Z 68.4K followers, 35.7K engagements

"http://x.com/i/article/2016734783876059139 http://x.com/i/article/2016734783876059139"
X Link 2026-01-29T04:48Z 68.3K followers, 10.4K engagements

"This paper on the current state of AI Agents is worth reading. Main points: [--]. Add memory to agents. [--]. Build agents as loops not pipelines. [--]. Go for RL only after the Agent's behavior is reliable. [--]. Specify which tool to use when (don't dump [--] tools into a prompt and hope for magic). [--]. Multi-agent systems fail unless the roles of planner executor and critic are explicit. ------------ Some of these lessons feel obvious but it's good to know that engineers from Meta Amazon and Google agree with each other. Cheers. https://arxiv.org/abs/2601.12538 https://arxiv.org/abs/2601.12538"
X Link 2026-01-29T06:35Z 68.5K followers, 36.4K engagements

"@navneet_rabdiya I understand. The content is something I have listed earlier I am very happy with the topics we are covering. This post was more about the practical benefits of a live program. Thanks for the comments :)"
X Link 2026-01-30T04:22Z 68.3K followers, [--] engagements

"The problem with hyping these websites is. People without a foundational understanding of AI panic. Skynet. The Matrix. AGI. What a load of nonsense. The models are randomly generating tokens and spewing it at each other. They aren't coming up with ideas. At best they are rambling in a space of speech possibilities. Any unique ideas Any sign of "awakening" No. People who are hyping these models are clutching at straws hoping to justify the ridiculous predictions of AGI they made earlier. No hate for Mr. Karpathy the tweet is for others who lose their minds at the slightest sign of novelty in"
X Link 2026-01-31T05:14Z 68.9K followers, 63.1K engagements

"AI Agents in [----]. Over the past couple of years the engineering around agents has been steadily improving. Some common agentic patterns: [--]. Monolith ❎ Microagents ✅ Like microservices software teams work on different agents independently. [--]. Perfect for Internal Tooling Dashboards admin panels etc. are low risk and can tolerate latency. Write the internal tools code (Front end APIs) using AI agents. [--]. Zero to One with Agents New feature Write an agent for it. If users show interest only then invest dev time in improving it. --------------- Developers were sceptical of AI Agents in [----] (and"
X Link 2026-02-03T20:23Z 68.4K followers, [---] engagements

Limited data mode. Full metrics available with subscription: lunarcrush.com/pricing