Vatsal Shah

This is a great article that covers the economics of GPUs, in terms of renting vs buying and understanding the components apart from a GPU that makes large scale machine learning possible! Do read if you have the time - https://lnkd.in/d3iNPs6C

14

Another notable thing from the new Budget: Space sector gets 1000 cr ($120M) VC Fund allocation. This is a very important signal, even though the numbers are small in the context of capital requirement by the sector. I am a big believer of India owning the space sector worldwide over time in a manner similar to how we have owned the Technology services market. The talent pool and infrastructure (thanks to ISRO), emerging technologies around miniaturization, and evolving supply chain makes it one of the few spaces where India has a ‘right to win’. Strongly recommend more Entrepreneurs to participate in this area. Given the high capital requirements in the initial days (before monetization), it is a hard place for VCs to play but we will see innovation in capital structure over time (the Government allocation being one such evolution). And I am not only saying this, I actually brought a company to our Investment Committee (where I am also a member) making a case for Investment. We continue to look at the space and hope to cross the line sometime. https://lnkd.in/gzB9UUb8

102

14 Comments

I believe a solution to this is currently being developed.

5

Final AMA for AIR5 For the @AntlerIndia Residency #5, apps from individual founders are due Aug 2nd. Existing teams can continue to apply (rolling decisions). Nandini Vishwanath & I are doing one final AMA this evening (July 29, 6-7 pm IST). We cover all things Residency - the best way to move from your -1 to 0 phase to an investment decision. https://lu.ma/7n35d1z5

42

2 Comments

𝗪𝗶𝗹𝗹 𝗔𝗜 𝗛𝗶𝘁 𝗮 𝗗𝗮𝘁𝗮 𝗪𝗮𝗹𝗹 𝗶𝗻 𝟮𝟬𝟯𝟬? 🆘 Very interesting analysis from epoch.ai about data corpus available for training large language models and whether we will hit the ceiling soon. https://lnkd.in/gD78f9DW The authors estimate that all available data will be used by ~2030, but they're also the first to admit that by then we will have ways to address "the shortage" (such as synthetic data, data efficiency improvements, use of other modalities). What I found more interesting though is why the latest predictions differ from their 2022 paper, where they speculated that all the data will be used by 2024: ☑ Web data proves more valuable than expected (on par with scientific papers) ☑ Models can "reuse" datasets multiple times 2030 is right around the corner, so we will see a lot of innovation on this front and creative ideas on how to 'recycle' and make best use of the data we already have. These estimates also don't take into account proprietary data owned by individual companies. In this No priors episode (https://lnkd.in/g2BkTJBz) there is a mention of JPMorgan owning 150PB of proprietary data, which is at least 150x of what was used to train GPT-4. JPMorgan is monetizing this data today (https://lnkd.in/gXjE92qx), but it gets exponentially more valuable in a world where data is a finite (and extremely sought after) resource.

12

A Delhi techie cleverly purchased the Jiohotstar domain before the Disney-Jio merger. What would you do if you found yourself holding a key piece of real estate in the domain world just as two giants are gearing up for a merger? This move highlights the intersection of foresight, opportunism, and digital ingenuity. It’s a masterclass in understanding the value of "virtual property" in today's internet-driven landscape. For business owners and tech enthusiasts, here's a thought: 1. Always be on the lookout for trends and upcoming mergers that could create opportunities in related domains. 2. Understand the power of a strategic digital asset. A seemingly simple domain name can hold substantial leverage. 3. As unique as the price might be, ensure to evaluate both short-term gains and long-term relationships before deciding to sell. It’s a fascinating time for tech and business alike as these unexpected opportunities pave the way for creative strategies #DigitalRealEstate #DomainInvestment #TechTrends #MergerOpportunities #Entrepreneurship #BusinessStrategy #Innovation #DigitalAssets #Foresight #Opportunism #InternetEconomy

5

Presenting my most detailed piece on AI covering applications & agents, which I believe are rapidly evolving as the biggest value investment opportunities for VCs in an industry of attrition amid expensive scaling laws. Full post - https://lnkd.in/gKYURDpA TL;DR (written by an AI application 😉) 🚀 The AI Economy: A Paradigm Shift The AI economy is generating tremendous value, driven by nascence, technological differentiation, commercial traction, and risks taken. Generative AI (GenAI) has led to the rise of not just chatbots like ChatGPT but copilots that enhance human workflows. ​ 💡 Investment Surge VCs invested $38.6 billion in AI and machine learning startups in the U.S. in the first half of 2024. ​ In Europe, €7.9 billion was invested in AI startups across 1,030 deals in YTD 2024. ​ 🔍 The GenAI Value Chain ​ Key categories: Chips, Compute, Models, Data, and Applications. ​ I make a case that the compute hyperscalers as well as the acceleration layer (together, the infra of AI) albeit are the biggest fish in the pond, would not be the purveyors of the best in AI agent and application innovation. 🤖 AI Agents: The Next Evolution AI agents can handle complex tasks with minimal intervention, from sales prospecting to scheduling meetings. ​ Fully autonomous agents, or digital coworkers, are still in early adoption phases. ​ These are nevertheless reinventing the way software was delivered and consumed. 📈 Applications & Agents: The Future of AI Applications are expected to drive the highest acceleration in AI. ​ While horizontal (& B2B) AI apps are gaining traction, vertical AI apps (esp. in regulated industries with a lot of data) could create strong moats. 💰 Monetization Strategies Direct monetization: Charging for AI features directly. ​ Indirect monetization: Integrating AI features into existing bundles without altering the price. ​ 🔮 The Road Ahead Agents are coalescing around unique GTM motions, and I expect them to become the new paradigm of software delivery and consumption beyond SaaS. ​ #AI #ArtificialIntelligence #GenAI #AIApplications #AIAgents #TechInnovation #VentureCapital #FutureOfWork #DigitalTransformation

5

An interesting article from Chip Huyen - on building a complete Generative AI Platform based on existing LLM. It starts from the simplest architecture and progressively add more components to achieve fully production-ready solution like on the image below. The article is available at: https://lnkd.in/dEcyHfTM

42

Lifting the Veil on AI's decision making, Vinay Kumar Sankarapu from Arya.ai will share about AI transparency that goes beyond the usual "black box" discussions. In today’s world of complex deep learning models, understanding why a model makes certain decisions is crucial—especially in mission-critical applications. He'll explore DL-Backtrace, a new model-agnostic technique designed to bring transparency to any deep learning model, whether it’s an LLM, computer vision, NLP, or tabular data model. What You’ll Learn: - The importance of explainability for mission-critical deep learning applications - How DL-Backtrace compare to existing techniques like SHAP, LIME, and GradCAM for models including LLMs (Llama 3.2), NLP (Bert), and CV (ResNet) - Real-world benchmarking results showing DL-Backtrace’s performance across varied deep learning architectures - Insights into the current limitations and future possibilities in scaling explainability for large models #ResponsibleAI MLOps World Generative AI World

7

For People who like to explore LLM and GPU power, here is a Made In India Gift Contact if you want to learn more Swadeep Singh Aidwise.ai #Ola #Aidwise

15

6 Comments

From IIT to Goldman Sachs to founding a Y Combinator-backed AI company—Sudipta’s entrepreneurial journey is filled with valuable lessons. In this unscripted conversation, I sit down with Sudipta Biswas, co-founder of Floworks, to dive into his story: 👉🏻 IIT Days: How his time at IIT Kharagpur shaped his entrepreneurial mindset and prepared him to take on bigger challenges. 👉🏻 From Goldman Sachs to Entrepreneurship: What drove him to leave a stable role at Goldman Sachs to launch Umpire’s View, and how the failure of that startup taught him resilience and the importance of pivoting. 👉🏻 Founding Floworks: Sudipta shares how he identified a real customer pain point, secured funding, and built Floworks, an AI-driven sales automation platform that’s solving key problems for businesses. 👉🏻 The Y Combinator Experience: How YC helped sharpen his vision and refine his product-market fit, turning Floworks into a success story. 👉🏻 The Future of AI: Sudipta’s take on how AI will evolve to tackle even more complex challenges, transforming industries and boosting productivity. Packed with actionable insights, from market research to securing funding, this episode is a must-watch for anyone interested in entrepreneurship, AI, and innovation. YouTube 👉🏻 https://lnkd.in/g5drCQaw Thank you, Srishti M. Manthan Sehgal, for all the support. #AIStartups #GoldmanSachs #IIT #YCombinator

70

3 Comments

Our vision at Niti AI is to empower the next billion (non) developers to build and ship products with AI! Whenever I reflect on ambitious goals like this, I'm reminded of the below quote from The Fountainhead by Ayn Rand. Hari Subramanian | Aravind R. | LoopX

15

Venture capital firm Peak XV Partners has about Rs 16,000 crore of “dry powder” ready for investment Just think of it, a lot of problems in the ecosystem will be solved So just be strong, focused and lean for now Sooner or later, things are going to change

15

Not long ago, I posted about the wave of innovation coming out of India. This piece dives deeper. DeepTech startups make up 70% of India-origin filings. The space is growing quickly, and I thought this provided helpful guidance for the next steps. #Innovation #India #DeepTech

3

7 steps in which OYO Rooms' became a National Giant In the fast-paced world of hospitality, few firms have had as significant an influence as OYO Rooms. Their rapid expansion shows the power of a well-executed digital strategy. Here's how OYO used technology to scale rapidly and successfully in seven steps: 1. Data-Driven Decisions: OYO used advanced analytics to understand market demands, customer preferences, and operational efficiencies. Their data-centric approach enabled them to make informed decisions, optimise pricing, and enhance customer experiences. 2. Technology Integration: OYO integrated cutting-edge technology into every aspect of their business. From booking platforms to customer service, their tech stack ensured seamless operations and superior service delivery. 3. Scalable Infrastructure: By building a scalable digital infrastructure, OYO was able to rapidly expand into new markets without compromising on service quality. Their robust backend systems supported a growing network of hotels and homes globally. 4. Mobile First Approach: Recognising the shift towards mobile, OYO developed a user-friendly app that allowed customers to book rooms, check in, and manage their stays effortlessly. This mobile-first strategy significantly enhanced user engagement and satisfaction. 5. Global Reach with Local Adaptation: OYO's digital strategy wasn't just about global expansion; it was also about local adaptation. They tailored their offerings to meet local needs and preferences, ensuring relevance in diverse markets. 6. Agile Marketing Campaigns: Leveraging digital marketing channels, OYO ran targeted campaigns to reach potential customers. Their agile marketing approach, powered by real-time data, allowed for quick adjustments and a maximised ROI. 7. Partnerships and Integrations: Strategic partnerships with online travel agencies and other platforms amplified OYO’s reach. Integrations with third-party services provided added value to customers, making OYO a preferred choice. - Enjoyed this? ♻️ Repost it to your network and follow Suvam Sahoo for more. Want help to scale your brand and digital presence as a founder? Book a free discovery call with me - send a 'hi' in dm :) #Hospitality #DigitalTransformation #TechInnovation #BusinessGrowth #OYORooms #MarketingStrategy #CustomerExperience #GlobalExpansion #Leadership #Innovation

4

Reason we need a solid framework that can validate LLM response before it is presented to end customer Till such time GenAI in enterprise will be mostly focussed on human in loop use cases with only some of it direct customer facing #generativeai #artificialintelligence

51

24 Comments

India’s Oyo, once valued at $10 billion, is finalizing new funding at a $2.5 billion valuation. This marks a steep decline from its $10 billion valuation in 2019. The startup, which has raised over $3 billion from backers like SoftBank, Peak XV Ventures, Lightspeed, Airbnb, and Microsoft, has struggled to attract institutional investors. In recent months, it has aggressively pitched to high-net-worth individuals after withdrawing its IPO application twice in the past four years. Once a red-hot Indian startup active in markets including the U.S. and Europe, Oyo has now scaled back its international presence. #venture #downround https://lnkd.in/gE2GjYnM

5

📣Insightful development within ByteDance and its AI Department reveals key learnings for both current organizations and upcoming tech start-ups. An internal breach laid bare a possible vulnerability in their security protocols, which indeed is a concern we must all share. 1️⃣ByteDance, the creators of the popular TikTok platform, recently witnessed a security breach tied to an intern's alleged insubordination. 2️⃣Fortunately, no online operations or commercial projects were negatively impacted, only their AI commercialisation efforts. This does not nullify the plausible risks tied to such situations in the future. 🤔 I foresee companies taking keen note of this situation and amping up the security protocols, specifically relating to their AI departments. There is an apparent need for vigilance and tighter controls at each stage of accessing and processing data. - Companies need to rethink their boarding and training strategies to provide interns and new hires with the correct know-how of their job roles; also, imbuing the importance of adhering strictly to codes of conduct and respecting intellectual property. - The right balance must be struck between offering access required for employees to execute their roles and ensuring the absolute security of the project. - Emphasizing the need for an AI Ethicist role within the teams could prevent such damaging incidents. These are individuals skilled in AI algorithms, knowledgeable in terms of ethics & laws, cognizant of deploying AI models responsibly. In a technology-driven era where data and Artificial Intelligence sit at the core of business, ensuring the paramount security of these operations remains non-negotiable. This incident serves as an elucidative illustration of the potential hazards that companies could face and the preventive measures needed. Remember 👉 Prevention is always better than cure. Remain equipped, stay vigilant! #ByteDance #ArtificialIntelligence #SoftwareSecurity #PreventiveControls #AITraining

2