Created a list of around 100 books that had major impact on me over the last two decades. These are also the books that I've re-read several times. buff.ly/2xaaDLd

This is such a brilliant idea.

No.1 💎

Very excited to share Genie 4, our incredible interactive world model, bringing together real-time physics, reasoning, and audio! Simulations within simulations, makes you wonder just how deep the rabbit hole goes... I have a feeling we will be finding out soon enough...

What tasks should AI be trusted with, and when should humans remain involved? Designing those decision rights is one of the defining challenges of the AI era. Researchers at the MIT Center for Information Systems Research have developed a framework to help. Learn more: bit.ly/4zed20M

Last 3 overs of the epic knock of Virat Kohli in the Ind Vs Pak match, with scintillating commentary by the ausie commentator! 🔥 nitter.siddz.com/Melbourne__82/status/2…

Here's my updated review of each personal agent: 1. Grok Bot Pros: A team of bots is the easiest to understand. Free X access is amazing for research. Voice mode and the new Pages feature are great. Cons: Email feature doesn't really work. 2. Muse Pros: Free, best consumer UX, most meme-worthy avatar, and business calling is amazing. Cons: Muse Wonder is good but Opus/Astra is better. 3. Claude Duet Pros: Most proactive out of the box. Feels polished and simple to use. Arguably the best model behind it. Cons: Missing some features in research preview, like being able to drive its browser. 4. ChatGPT Dots Pros: Works inside ChatGPT, so it can drive all your cloud and local threads. Cons: Much better than at launch but still buggy. Voice mode is slow to connect. I've also tried Instinct and some of the other agents, but I don't want more than 4 at a time connected to my email. 📌 Don't miss my full Claude Duet review here for everything that it can do: piped.video/D-pz8L1orVI

Anthropic hired this engineer at $350K–$850K a year because he knows how to build harnesses for multi-agent systems In a 12-minute workshop, he shows exactly how to build one from scratch AI → Agents → Harness → Loops → Graphs step 1 → start with the Claude Agent → the harness handles loops, context, and sandboxing step 2 → separate the brain from the hands → reasoning in one place, tools in a sandbox, 60% faster to first token step 3 → run it server-side and log every step → close your laptop and it keeps running, crashes resume from the log step 4 → make failure cheap → retry dead sandboxes and replay lost context instead of starting over step 5 → turn yesterday's logs into new memory and skills → the harness wakes up smarter Anthropic calls this "dreaming" Most people spend weeks building this by hand You don't have to Bookmark and watch it Then read the full Opus 5.5 harness engineering guide below ↓

Remember Kahneman's Thinking, Fast and Slow? The brain decides when to rely on learned patterns (sys1-fast/Cerebellum + distr. networks) vs. deliberate thinking (sys2-slow/Prefrontal Cortex). Energy and compute constraints necessitates efficiency. Where is the System 1 for RL?

Ben Horowitz on treating your story as your strategy, and why founders often skip writing it even though it's one of the most important parts of the job: "The story of the company is also the strategy of the company... In your mind, you're kind of adjusting that strategy a little bit at a time every day. But then a quarter later, it's actually pretty different than what you started with." "So it's really important that you keep upgrading your story. The articulation of your story. I would definitely recommend it be in written form." "Something that you can share with everybody in the company, and everybody you're recruiting, everybody you're trying to raise money from, all your constituents, every customer." "The problem that I think founders have is it doesn't feel like work. Like, 'What are you doing today?' 'I'm working on my story.' That doesn't sound like work." "Because it doesn't feel like work, people don't do it, but it's probably one of the most important parts of the job." @bhorowitz w/ @speedrun GP @Tocelot

Must-read paper from Sakana AI on orchestrating coding agents for scientific research. (bookmark it) This is an important read with lots of useful insights if you are building multi-agent systems. I like this work because it presents a concrete multi-agent framework and an end-to-end distributed pipeline that lets human researchers conduct computational research alongside modern coding agents. What they did: A development version of their system helped write a paper that was accepted as a main-conference talk at ALIFE 2026, after human verification and edits. How they did it: The Agentic Scientist splits a project into stages. Task drafting, ideation, planning, sandboxed experiments, baselines and ablations, analysis, and report writing each run on a coding agent that checks its own work. A meta-controller runs many of these agents in parallel and checks every stage against a rubric. The ablations show that reliable autonomy depends on that orchestration layer as well as on the base model. arxiv.org/abs/2610.18933 Chat with Paper: academy.dair.ai/papers/the-a…

Argon has this amazing natural understanding of the world. It can do so many things from first principles. Here's a pelican riding a bike, using fourier epicycles. Video made by Argon. Time-synced Lyria 3.5 music prompted by Argon.

Google Brain founder Andrew Ng: "Prompting will be dead in 6 months Agent harnesses built with loops and graphs will replace it" Agent → Harness → Feedback → Loops → Graphs → Self-Improving Systems In this 1-hour Stanford lecture, he explains what the best engineers are building instead and how you can start today Prompt → Run → Verify → Improve The first 20 minutes teach what most $1,500 courses try to sell you For free Bookmark and watch it today

hoy leí esta guía sobre Harness Engineering que me pareció muy buena. explica cómo se diseñan y construyen agentes de IA, cubriendo temas como: - gestión del contexto y memoria - tools y MCP - orquestación de agentes - seguridad y permisos - validación, observabilidad y manejo de errores muy buen recurso para entender todo lo que hay detrás de agentes como Claude Code o Codex. documento: drive.google.com/file/d/1PT-… además, les dejo un artículo que escribí sobre Harness Engineering para complementar y seguir profundizando.

Naresh Jain compares coding agents with and without a harness. He found it made a frontier model more expensive in tokens for little gain, but make a weaker (and cheaper) model more effective. martinfowler.com/articles/ar…

🇮🇳 199 Indian startups have raised $10M+ in 2026. Here's every one: 1. CRED: $900M Series H ‡ 2. Neysa: $600M Series B 3. Sarvam AI: $300M Series B 4. KreditBee: $280M Series E ‡ 5. Rapido: $240M Series F 6. Simple Energy: $206.3M † * 7. Emergent: $200M † § 8. Spinny: $170M Series G 9. Udaan: $160M * ‡ 10. River Mobility: $120M Series C * 11. Ultrahuman: $118M † * 12. Navi: $100M 13. Slice: $100M ‡ 14. Pixxel: $100M Series C 15. Square Yards: $95M * 16. Yulu: $93M Series C * 17. Neo Group: $89.7M † 18. Ultraviolette: $85M Series E 19. Arya ag: $80.6M Series D ‡ 20. Ema: $77M Series B § 21. Freehand: $75M § 22. GPS Renewables: $66.4M Series C 23. Scapia: $63M 24. Swish: $62M † * 25. Rocketlane: $60M Series C 26. Skyroot: $60M Growth Round 27. Snabbit: $56M Series D 28. Popo Global: $56M 29. FirstClub: $55M Series B 30. Temple: $54M Seed 31. SolarSquare: $53M Series C 32. IDfy: $52.5M Series F ‡ 33. Zetwerk: $52M Pre-IPO 34. The Whole Truth: $51M Series D 35. Juspay: $50M Series D 36. Nua: $50M Series C 37. Euler Motors: $47M Series E 38. cult fit: $47M Series G 39. Pronto: $45M Series B 40. Noon: $44M § 41. Third Wave Coffee: $43M 42. Ecofy: $42M 43. Flam: $40M Series B 44. Mozark: $40M Series B § 45. Palmonas: $40M Series B 46. Raghu Vamsi Aerospace: $40M 47. Veriqus: $40M 48. BusinessNext: $40M 49. Vetic: $40M 50. BookMyShow: $40M (reported) 51. upGrad: $38M 52. Airbound: $37M Series A 53. Atlys: $36M Series C 54. Yuma Energy: $35M Series A 55. DailyObjects: $34.3M Series C ‡ 56. Sahi: $33M Series B 57. Pee Safe: $32M Series C 58. Fast&Up: $32M Series D 59. Hang Ten Systems: $32M Seed § 60. Rozana: $31.6M Series B 61. Sukino: $31M Series B 62. Orange Health Labs: $30M Series C 63. Equal: $30M Series B 64. Innefu Labs: $30M Series B 65. Supertails: $30M Series C 66. Emversity: $30M Series A 67. Centricity: $30M * 68. TruNativ: $30M Series B 69. Easy Home Finance: $30M Series C 70. Limelight Diamonds: $29.5M 71. Ethereal Machines: $28.5M Series B 72. Wint Wealth: $28M Series B 73. Unbox Robotics: $28M Series B 74. InRisk Labs: $27M Series A 75. Pramaana Labs: $27M Seed 76. Deccan AI: $25M Series A 77. Stable Money: $25M Pre-Series C 78. BillionE Mobility: $25M * 79. Matter: $25M 80. Assiduus Global: $25M Pre-Series B 81. Escape Plan: $25M Series A 82. Arboreal Bioinnovations: $24M Series A 83. MyGate: $23.6M 84. Knight Fintech: $23.6M Series A 85. Agilitas Sports: $23.5M 86. Olyv: $23M 87. Recykal: $23M Series D 88. WayCool: $22.7M 89. NextGen Finance: $22.6M Series A 90. Kimbal Technologies: $22M Series B 91. Nava: $22M Series A 92. Mosaic Wellness: $21.7M 93. Jain Cord: $21.5M Series A 94. Tsecond ai: $21.5M 95. AllHome: $21.2M Series B 96. Exponent Energy: $21.1M Series B2 97. QNu Labs: $21M Series A 98. Runable: $21M Series A 99. AssetPlus: $21M 100. Plum: $20.5M Series B 101. EtherealX: $20.5M Series A 102. Even Healthcare: $20M 103. Varaha: $20M Series B 104. Constelli: $20M Series A 105. NeoGeo: $20M Series A 106. Nexedge Capital: $20M 107. Bellatrix Aerospace: $20M Pre-Series B 108. Liquidnitro Games: $19.1M Series A 109. Blue Tokai: $19M Series D2 110. Pandorum Technologies: $18M Series B 111. Statiq: $18M 112. Mokobara: $18M Series C 113. Elevate Education: $17.7M Series D 114. Wishlink: $17.5M Series B 115. The Indus Valley: $17M 116. JustAI: $17M Series A 117. BlissClub: $16.8M 118. Xflow: $16.6M Series A 119. Showroom B2B: $16.5M Series A * 120. The Hosteller: $16M Series B 121. Anveshan: $15.8M Series B 122. Incuspaze: $15.6M 123. ZILO: $15.3M Series A 124. SUGAR Cosmetics: $15.3M 125. Plazza: $15M Series A 126. GobbleCube: $15M Series A 127. Fairdeal Market: $15M Series A 128. C2i Semiconductors: $15M Series A 129. Portkey: $15M Series A 130. Vorflux: $15M 131. Whizzo: $15M Series A 132. Mitigata: $15M Series B 133. Gravity: $15M * 134. Zenskar: $15M Series A 135. Piston: $15M Series A 136. Transvolt Mobility: $15M 137. Aerem Solutions: $15M Pre-Series B 138. Bonkers Corner: $15M Series A 139. Loop AI: $14M Series A 140. Quick Clean: $14M Series B 141. Oolka: $14M Series A 142. Matel Motion & Energy: $13.6M Series B 143. Graph AI: $13.3M Series A 144. Captain Fresh: $13.3M 145. Aerchain: $13M Series A 146. StrainX Bioworks: $13M 147. HrdWyr: $13M Series A 148. Smallest ai: $13M Series A 149. Apna Mart: $12.7M Series C 150. Wingreens Farms: $12.6M Series D 151. Rezolv: $12.5M Series A 152. Coral: $12.5M Seed § 153. Bachatt: $12M Series A 154. Prayaan Capital: $12M Series A 155. InstaAstro: $12M Series A 156. Cradlewise: $12M Series A 157. HomeRun: $12M Series A 158. Optimist: $12M Pre-Series A 159. Lumio: $12M Series A 160. GIVA: $12M Series C 161. Naturis Cosmetics: $11.8M Series A 162. BGauss: $11.6M Series D 163. Innoviti: $11.4M Series M 164. Reo Dev: $11.3M Series A 165. RISA Labs: $11.1M Series A 166. KaarTech: $11M Series B 167. LightFury Games: $11M Pre-Series A 168. StrideOne: $11M 169. Dhruva Space: $11M 170. BatX Energies: $11M Series A 171. Off Beat: $10.7M Seed 172. Pluckk: $10.7M Series A 173. HOCCO: $10.7M 174. Aham Housing Finance: $10.7M 175. Rusk Media: $10.6M Pre-Series C 176. Comet: $10.6M Series B 177. Immuneel Therapeutics: $10.5M Series B 178. HyugaLife: $10.5M Series A 179. Bakingo: $10.5M Series B 180. Credilio: $10.5M Series A * 181. Rivet: $10.5M Seed 182. Beyond Appliances: $10.4M Series B 183. Balwaan Krishi: $10.4M Series B 184. Seeds Fincap: $10.4M Series B 185. E3 Electric: $10.4M Series A 186. Inspecity: $10.4M Pre-Series A 187. StockGro: $10.4M Series B 188. Arivihan: $10.2M Series A 189. Material Depot: $10M Series A 190. Vervesemi: $10M Series A 191. CloudSEK: $10M Series B 192. Kapture CX: $10M Pre-Series B 193. TrueFan AI: $10M Series A 194. Gnani ai: $10M Series B 195. Quanfluence: $10M 196. Carrum Mobility: $10M Series B 197. CARPL ai: $10M Series A 198. MeltPlan: $10M 199. Ringg AI: $10M Series A † Repeat raisers: 5 startups raised twice this year. Their amounts are combined into one line: - Simple Energy ($180M + $26.3M) - Emergent ($130M + $70M) - Ultrahuman ($70M + $48M) - Neo Group ($53.4M + $36.3M) - Swish ($38M + $24M) * Includes debt alongside equity ‡ Headline figure includes secondary share sales (or, for Udaan, conversion of existing bonds), not only fresh capital § Headquartered outside India (US or Singapore), founded by Indians with teams in India Period: Jan 1 to Oct 12, 2026 Sources: Inc42, Entrackr, press releases

Instead of just watching an hour of Netflix, why not dive into this 2-hour Stanford lecture on AI careers? It’s packed with insights that could give you a real edge in the AI race, more than all the AI content you’ve seen this year.

My Letter to the People of Mysuru Dear Mysureans, Namaste! Last Sunday, I was in Mysuru to participate in the L’Étape Mysuru cycling event. Congratulations on having such a wonderful cycling event find its home in your beautiful city! The event was fantastic and very well organised. But what struck me most was just how beautiful it is to cycle in Mysuru, inarguably Karnataka’s most beautiful city. The wide roads, massive green trees, cool morning breeze, the challenging climb up Chamundi Hill and the rewarding downhill, the beautiful roads through quaint villages towards Varuna. Every bit was absolutely beautiful. Many fellow cyclists who have cycled extensively abroad remarked that the experience was comparable to, and sometimes even better than, cycling in Western countries. It is precisely for this reason that I am writing to you. Please protect what you have. Do not allow this beautiful city to be destroyed by stupid, unplanned urbanisation that eats up everything beautiful, aesthetic and life-giving about a city. Unfortunately, this has happened to a great extent in Bengaluru. I have four humble requests. First, please do not allow flyovers inside your city. Flyovers are among the ugliest things we can do to a beautiful city. They do not sustainably solve traffic congestion. Instead, they create ugly concrete monsters that destroy the aesthetics and character of a city. If anyone proposes flyovers as a solution to Mysuru’s traffic problems, reject such proposals outright. Do not allow your beautiful boulevards to become concrete jungles. Second, demand a completely underground Metro system for Mysuru. Mysuru will grow. Economic activity will increase, and so will the demand for better urban mobility. Please start planning for this future today. An underground Metro can become the backbone of Mysuru’s public transport while preserving its beauty and aesthetics. Yes, it may be more expensive initially, but the long-term benefits to the city’s character and quality of life will make it worth every rupee. Third, protect your boulevards, trees, parks, lakes and open spaces at all costs. One reason urban life in most Indian cities has become so stressful is the grave shortage of green and open spaces. These are not luxuries. They are essential to our physical and mental well-being. They allow us to slow down, breathe and reconnect with our real home, nature. Mysuru still has so much of this. Please do not allow these treasures to disappear in the name of development. Once lost, they are almost impossible to recreate. Fourth, preserve Mysuru’s walking culture. One of the most heartening things I noticed was how many people still walk long distances. This wonderful culture has almost disappeared from many cities because we have started building cities exclusively for cars, not people. Invest heavily in good footpaths, pedestrian crossings, shaded walking streets and safe public spaces. Make walking the most natural and enjoyable way of moving around Mysuru. Along with an underground Metro and good public transport, this can make Mysuru a model for Indian urban development. Mysuru is not just another city. It is a cultural symbol of Karnataka. Its history, heritage, institutions and contribution to modern, progressive Karnataka make it truly special. Bengaluru has unfortunately made many mistakes in its urbanisation. Mysuru has the opportunity to learn from them rather than repeat them. Development need not come at the cost of beauty. Modernisation need not mean destroying heritage. Economic growth need not mean making cities unliveable. I sincerely hope Mysureans will take an active interest in how their city grows, support good urban planning, oppose bad infrastructure proposals and protect everything that makes Mysuru special. You have inherited one of India’s most beautiful cities. Please ensure that future generations inherit an even more beautiful Mysuru. Love, Tejasvi Surya

Charlie Munger on the secret to life. “I just stay away from the problems that can’t be fixed and pick the ones that can." "I don’t like unlimited failure. I don’t want to fish forever and never catch a fish. I have to have some reinforcement. And so I pick some things that can be done and do them." "But I do think that if you’re reasonably obsessed with something, even if it’s intermittent, and you have a long attention span, you keep working over the serious problems, that you’ll stumble into an answer. That’s half the secret of life.” — Charlie Munger

What tasks should AI be trusted with, and when should humans remain involved? Designing those decision rights is one of the defining challenges of the AI era. Researchers at @MIT_CISR have developed a framework to help. Learn more: bit.ly/4zed20M

IMO the only tests you should have in the age of coding agents are, in order of priority: 1. Full E2E tests. Nothing mocked out at all. Can even be something that runs in prod on test accounts via Playwright or equivalent. 2. Integration tests. Agents make more mistakes as things bleed between data/API boundaries where schemas can drift. 3. Golden tests. This helps ground the code with real examples of data, and can be used as regressions against edge cases. Unit tests are bloat 99% of the time if you’re using a frontier model since they’re smart enough now to get the implementation (and subsequent iterations) right on the first shot, so they only add bloat at this point.