Dilip Ramesh

Another big ed-tech ‘Unacademy’ is up for sale, merger or bailout. Company fired 250 of their employees yesterday and is struggling with finances. It has approached the likes of Allen, PhysicsWallah, K12 Techno and others. It was also in talks with K12 Techno about the merger but looks like those talks are not advancing. Company has raised $877 million till date and company is searching for a respectable way out. Unacademy's situation highlights the potential turbulence within the EdTech sector. While the industry has seen immense growth, companies are now facing increased competition and the need for profitability. What are your views?

10

My recent experience with the Bihar startup ecosystem paints a frustrating picture, highlighted by a news headline: "Worst decision of my life: Bihar-based semiconductor startup's." This reflects the struggles entrepreneurs face in a state that claims to support startups but fails in execution. One such failure occurred when I emailed the District Magistrate of Darbhanga and Smt. Bandana Preyashi, IAS, through the official startup support email, only to discover the email service wasn’t functional. This inefficiency speaks volumes about the broken system that is meant to foster entrepreneurship. Even direct emails to the DM and the Secretary went unanswered, suggesting that the officials tasked with fostering innovation are not accessible or engaged. This disheartening experience raises questions about whether these roles serve any real purpose or are just symbolic. The message to aspiring entrepreneurs is clear: the system is unresponsive, and progress seems impossible. The Bihar government has developed a startup policy, but policies alone won’t build an ecosystem. There needs to be infrastructure, accountability, and effective communication channels. Ministers like Nitish Mishra and Jitan Ram Manjhi may have good intentions, but their efforts are not producing real results. As global startup ecosystems rapidly evolve, Bihar seems stuck in the past. Without incubators, accelerators, and tangible support, Bihar’s dream of becoming a startup hub will remain just that—a dream. #Bihar #Startups #Entrepreneurship #GovernmentFailure #PolicyIssues #InnovationCrisis #StartupEcosystem #LeadershipFailure #Inefficiency #BiharDevelopment #EmptyPromises #StartupStruggles #EntrepreneurshipChallenges #AccountabilityNeeded #BrokenSystem #PoliticalPriorities #FailedSupport #BiharStartups #EntrepreneurialFrustration #BusinessChallenges #PolicyWithoutAction #GovernmentInaction #BiharPolitics #InnovationNeedsSupport #DreamsVsReality #StartupHubDreams #MissedOpportunities #LackOfInfrastructure #InnovationStuck #BihariTalent #SupportForStartups

34

11 Comments

Discover Mockingbird, Vectara’s advanced Retrieval Augmented Generation #RAG and structured output-focused Language Learning Model #LLM. Suleman Kazi, Vivek Sourabh, Rogger Luo & Abhilasha Lodha take you on a technical deep dive into Mockingbird's impressive performance and capabilities. 1️⃣ Why Another LLM? - Learn why Vectara developed its own LLM focused on tasks that matter to our customers. 2️⃣ Training Mockingbird - Explore the rigorous training process for RAG and structured output tasks. 3️⃣ Evaluation Metrics - See how Mockingbird excels in generation quality, citations, and structured output. 4️⃣ Human Ratings - Understand the human evaluation process and how Mockingbird compares to leading models like GPT-4. Mockingbird is integrated into the Vectara platform, offering unparalleled performance with high-quality, secure outputs. 🌐 Dive into the full blog and see how Mockingbird can transform your data handling and generation processes. https://gag.gl/2WqJGM?

42

Judgemental Vs Enabling conversation. We all swing between these modes, but it is our responsibility to catch hold of ourselves and keep moving the conversation into enabling mode all the time. 𝗦𝗰𝗲𝗻𝗮𝗿𝗶𝗼 𝟭 Q - How is the plan Vs actual for yesterday ? A - We are at 85%. Q - Why A - Reasons1 , Reason 2 ..... The conversation is closed with an agreement the team member is not capable of handling the deliverables and is always giving reasons. Do we see these kind of scenarios around us ? 𝗦𝗰𝗲𝗻𝗮𝗿𝗶𝗼 𝟮 Q - How is the plan Vs actual for yesterday ? A - We are at 85%. Q - Why A - Reasons1 , Reason 2 ..... OK, Let's work on how to remove these reasons and what help do you need to improve the plan vs actual numbers. What help do you need ? What do you think out the above scenario will generate power to remove the reasons and help to improve the performance of the teams ? Comments 👇 Amarpreet Singh

11

2 Comments

I'm really glad the Kolkata r*pe case is getting the attention it deserves. It started with doctors standing up for one of their own and quickly became too big to ignore. But even now, the facts and discussions around them are scattered across social media. It would be better if everything happened in one place. Sadly, many similar incidents go unnoticed every day. Even with social media, you mostly see what big celebrities or people you follow share. Not every victim has a big group or community to amplify their voice. This is why I started WeVoice News. Some victims don't have access to the same reach, and their stories often go unheard unless someone with a big following speaks up. WeVoice is here to change that by putting the power of distribution in everyone's hands. We’ll highlight all the important stories and let you boost what you feel matters. Basic human rights violations like what happened in Kolkata, corruption, environmental issues, infrastructure, health, education—every repeating problem deserves its voice and space. We're working hard to make this a reality. Stick around, follow us, and help us build this community. We're doing our best too. WeVoice News

27

"The average disposable income of Bihar is lower than Bangalore etc. Don’t you think that will play into what sort of QSR should be built in Bihar right now?" I got this question recently in response to my recommendation for considering an investment into Cafe Hideout, who in my mind are building a land-grab in Patna and can perhaps dominate eastern India in next 3-5 years. I was reminded of Rory Sutherland's perspective of the horizontal and the vertical way to look at data. Clearly, BLR has a much higher average disposable income than Patna (and most parts of eastern India). No doubt about it. But simply looking at the average misses the skewed data from Patna and other eastern states. Patna has about half a million folks with disposable income higher than BLR imo and abou 5 million folks who are at the bottom of the per capita GDP table. If we only look at the average, we will miss the opportunity to think of the needs of the top half million (which is a big number) completely. BTW, their aspiration is to have a lifestyle which matches the top 0.5% of folks in T1 cities. Aspirations are changing everywhere in Bihar as the Digital Desh team has discovered (you can read their insights and research at https://lnkd.in/gYFyk7BE. The report is free to download. Thanks Jasminder Singh Gulati for the coverage). Bihar is changing. And boy does it have size to turn on the momentum.

61

5 Comments

🎟️ Coldplay is coming to Mumbai, and tickets are about to go live on BookMyShow! 🎶 As I’m figuring out how BookMyShow will manage the queue to ensure fairness (and how I can maximize my chances of grabbing those tickets 🏃‍♂️💨), I started thinking about what’s happening behind the scenes. So, while preparing for this “Fastest Finger First” war, here are my thoughts on how BookMyShow might be gearing up for the massive ticket surge: 1️⃣ Virtual Waiting Rooms: To prevent crashes, users are placed into a FIFO (First-In, First-Out) distributed queue. But here's the catch: how do they actually decide who came first? 2️⃣ Precise Timestamps: Each click to "book" is timestamped down to the millisecond (or even nanosecond?). But speed also depends on how fast your request reaches their server! 👉 The kicker: Even if your request is lightning-fast, it may not be the first to hit the server depending on your location. If you’re in Bangalore and someone else is in Mumbai, and the servers are based in Mumbai, their request might reach faster—even if you both clicked at the same time. 🌍 Example: The network latency (round-trip time) between Bangalore and Mumbai is typically 30-40 milliseconds, while for someone in Mumbai, it could be just 5-10 milliseconds. So, if both users click at exactly the same moment, the person in Mumbai has an edge, with their request hitting the server faster by 20-30 milliseconds! In a high-stakes scenario like Coldplay ticket booking, that small difference could mean the difference between getting tickets or missing out! 🥲 3️⃣ Centralized Queue Management: Once you hit the button, your request is sent to a centralized queue (likely using Kafka, Redis, or RabbitMQ). Even if requests come from different servers, the queue maintains global order based on timestamps—so the faster your request hits, the higher up you’ll be. 4️⃣ Concurrency Control & Atomic Operations: With thousands of clicks pouring in, atomic operations (e.g., Redis INCR) ensure no two users claim the same spot. The system processes one request at a time for fairness. 5️⃣ Race Conditions: If two users click at the same millisecond and their requests reach the server at the same time, tie-breakers like randomization might decide who gets ahead. 6️⃣ Scalable Infrastructure: Auto-scaling servers are ready to handle any load, whether it’s 10,000 or 1 million users. 7️⃣ Rate Limiting & Bot Protection: Real fans get priority over bots thanks to rate limiting and DDoS protection, ensuring fairness. 8️⃣ Real-Time Queue Updates: While you wait, WebSockets or Server-Sent Events (SSE) keep you updated on your queue position and other information like how many seats are left, so you know exactly where you stand. It’s fascinating how distributed systems, network latency, and queuing mechanisms shape events like these. Sometimes, it’s not just about speed!

24

Continuing with our guesstimate series, in the comments section of the last post, Sumant Malhotra asked, "Estimate the number of EVMs required in Bangalore for upcoming elections?" 🙌🏼 So, presenting the solution below, swipe to see the answer. 😎 Do let me know in the comments section, what approach you would take to solve the same problem. 🤔 #product #interview #productmanagement #guesstimates

31

6 Comments

Rajan asked Sam Altman, a year back - "how can Indians build an AI model like ChatGPT", and it was laughed at that time! We are changing that narrative - we are building AI for India, that is more human than humans. Our first model is Ira - she speaks Hinglish, and is the most amazing conversational AI in the world currently. Try her out - https://ira.rumik.ai

57

8 Comments

The Titan Caratlane acq at EV OF 17000Cr and founders 27% equity bought for 4621Cr is another shining example of how determined but ever learning founders turned cash burning biz model into profitable one and made one of the biggest cash exits in India. Another inspiring story for next gen founders. Aviral Bhatnagar Mithun Sacheti https://lnkd.in/d9rt7wGQ! The company struggled. Losses were mounting. VCs ignored them. For nearly 4 years, they trudged on while trying to make use of whatever money they had. So when CaratLane needed more money in 2019, Sacheti was in for a rude surprise. He expected Titan to give him the cash at a multiple similar to the previous round — 5 times his sales figure. But nope that didn’t happen. Titan turned around and asked, “So, what about the profits?” As if that wasn’t enough, the legendary Rakesh Jhunjhunwala, who was also an investor in Titan, asked — “Where is your cashflow?” And looks like that was the wake up call CaratLane needed. They tweaked their strategy. Instead of spending heavily on opening new stores, they went the franchise route. They snagged working capital funding at a low cost by leveraging the Titan partnership too. And finally, in FY21, CaratLane reported its first profits. Things were falling into place. The company nearly doubled its store size within the next year. And the profits swelled even further.

4

1 Comment

At IIT Mandi, reincarnation, out-of-body experiences are now compulsory topics for BTech students. Pseudoscience FTW. You may know the IIT Mandi director from such hits as "eating meat causes landslides" and "ghosts are real, I drove one away.” I've noticed a rising trend to desperately prove that all ideas of ancient India had a scientific basis, and to discredit evidence-based science as western. If it's indeed so scientific, let’s subject Indian Knowledge Systems to the same peer-reviewed, double-blind studies that form the basis for scientific thought. If not, then such ideas are just an exercise in hubris. Indian Institute of Technology, Mandi

278

78 Comments

1. A crore is still a crore regardless of how big the company is. You’re still answerable to shareholders. 2. Even if they had a crore to spare, this would only encourage others to do the same and make a quick buck. 3. Domain squatting, while is a grey area in India, is illegal in many countries and is unethical regardless. 4. Buying a domain (and possibly many others as a backup) is not an entrepreneurial skill. It costs barely ₹300-400 to buy a domain these days.

8

The State government of Delhi is thinking about cloud seeding to make it rain, but there’s only a 30-40% chance it will work. Delhi needs long-term fixes for pollution, not quick hacks like this. Plus, Delhi’s busy skies ( Air traffic) make it harder to pull off. The state government is blaming the central government for failing to create artificial rain, trying to distract from its own 10 years of failure in tackling the pollution crisis.

208

19 Comments

Koo India's homegrown social media platform, has announced its closure. Despite reaching 50 million downloads and integrating AI with ChatGPT, financial pressures and fierce competition proved too challenging. Co-founders Aprameya Radhakrishna and Mayank Bidawatka cited unsustainable costs and failed partnership talks. Read here to know more: https://lnkd.in/g-F6CCgQ Koo's highlights: Language-Based Microblogging: Catering to diverse Indian audiences. AI Integration: Empowering users with innovative tools. Governmental Endorsements: Winning the AatmaNirbhar Bharat App Innovation Challenge. Koo's journey underscores the tough road for homegrown platforms in the digital age. #Koo #SocialMedia #TechNews #Innovation #DigitalIndia

17

How much AI is too much? This is a perfect example of how topping AI over everything is undermining the power of wonderful tool that can create value with the right balance. The key to unlocking the true potential of AI lies in collaboration with human strengths. #HumanAI isn't about who's better, it's about working together for a brighter future. What are your thoughts on the human-AI balance? Let's discuss! #humaaiinteraction #AIAssitant #GenerativeAi #HumanAI

5

While IPL over the years have become bigger and buzzier, team Enqurious (formerly Mentorskool) has been busy thinking what if BCCI comes to us tomorrow to build an application to capture, store, clean, build and deliver data driven insights for IPL, how would we go about it? Well, the first step towards this is to model the real world in the data world. And that's where comes a "data model". Sorry, ML folks this is a model different that what you build :) Here's an ERD we developed as a part of our Data Modeling Learning experience based on the most popular tournament of the country. Teaching a dry topic like Data Modeling via cricket has been a game changer for us helping us engage the GenZ learners fast. Check out more scenarios across the data stack at our website : https://lnkd.in/gJjRGCak At Enqurious (formerly Mentorskool) we are a team of Data practitioners (Ex. Data Architects, Data Modelers, Data Engineers, Data Analysts and Data Scientists & AI Researchers) who also happen to love teaching and building near real world use cases from grounds up and then apply instruction design principles to develop learning experiences with integrated playground labs across job personas in the data + AI space. Whether you are in individual Data practitioner, leading a data team or lead the L&D dept of a fast paced analytics driven firm, Having data warehouse design and implement skills are a must! #DES2024 and check out our customizable, industry aligned learning solutions fitting the needs of your data team. #experientiallearning #datamodelling #ERD #entity #relations #cardinality

16

Naarica has been smashing it 💯 So proud of how far Shruti Chand has come. When we first met, she didn’t even have a website, just some traction from some offline events. And now, she’s doing Rs. 10L+ MRR with a double digit EBITDA margin. And she’s just touched the surface - the real mission is 1000x larger than this. In India 🇮🇳, there are still 12-15 crore women who use sub-par products (or none at all) during periods. Making women feel more confident, lesser anxious - I feel that’s Naarica’s true mission. A truly women centric brand. I hope there are more and more folks like Shruti who take up such bold challenges. I hope more and more folks get inspired by her and choose entrepreneurship. And truly solve India-specific problems 🙌🏽

122

5 Comments