flâneur

Daniel Kiss

113 followers · 255 following · 43833 views

on the atlas — 242

highlights — 904

  • “You want to learn expertise in one area and bring it to another area where it’s useful but not yet used,” he said. “That’s the golden goose.”
    Jacob Tsimerman Wins 2026 Fields Medal for André-Oort Conjecture Proof | Quanta Magazine
  • “Two years ago, I knew I was close, and I did let myself become obsessed,” he said.
    Jacob Tsimerman Wins 2026 Fields Medal for André-Oort Conjecture Proof | Quanta Magazine
  • Work on interaction models and “guardian angels” (models which are actually aligned to and customised for a particular user).
    What we’d like to fund — Paradigm 3
  • A friend of his named Jacob Chi said, “I take him to a party sometimes. He doesn’t talk, he’s absorbing everybody. I say, ‘There’s a human decency; you must talk to people, please.’ He says, ‘I enjoy your conversation.’ Six months later, he can say who sat where and who started a conversation, and he can repeat what they said.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • Another Harvard professor, Barry Mazur, told me that he was “moved by his intensity and how brave and independent he seemed to be.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • Probably most mathematicians are very low-key about getting a prize, because you’re not in it for the prize, but he seemed particularly low-key. It didn’t seem to affect him at all.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • In it, Peter Sarnak, a member of the Institute for Advanced Study, says that one day he ran into Zhang and said hello, and Zhang said hello, then Zhang said that it was the first word he’d spoken to anyone in ten days. Sarnak thought that was excessive, even for a mathematician, and he invited Zhang to have lunch once a week.
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • According to Deane Yang, a professor of mathematics at the New York University Polytechnic School of Engineering, a mathematician at the beginning of a difficult problem is “trying to maneuver his way into a maze. When you try to prove a theorem, you can almost be totally lost to knowing exactly where you want to go. Often, when you find your way, it happens in a moment, then you live to do it again.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • “I was tired,” he said. “But many times I just feel peaceful. I like to walk and think. This is my way. My wife would see me and say, ‘What are you doing?’ I said, ‘I’m working, I’m thinking.’ She didn’t understand. She said, ‘What do you mean?’ ” The problem was so complicated, he said, that “I had no way to tell her.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • “You must have been unhappy.” He shrugged. “My life is not always easy,” he said.
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • “When I looked into his eyes, I found a disturbing soul, a burning bush, an explorer who wanted to reach the North Pole.”
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • The British mathematician G. H. Hardy wrote in 1940 that mathematics is, of “all the arts and sciences, the most austere and the most remote.” Bertrand Russell called it a refuge from “the dreary exile of the actual world.” Hardy believed emphatically in the precise aesthetics of math. A mathematical proof, such as Zhang produced, “should resemble a simple and clear-cut constellation,” he wrote, “not a scattered cluster in the Milky Way.” Edward Frenkel, a math professor at the University of California, Berkeley, says Zhang’s proof has “a renaissance beauty,” meaning that though it is deeply c…
    Yitang Zhang’s Pursuit of Beauty in Math | The New Yorker
  • She first backed it in 2019, three years before ChatGPT. Inference is what AI does once it has been trained. If a model is a lawyer, training is law school and inference is the job itself: reviewing contracts, depositions, generally doing what a lawyer was taught. Every Claude query is inference. Every Harvey contract markup is inference. Every Tesla rerouted to a Whole Foods for magnesium supplements is inference.
    Sarah Guo's Wager - Colossus
  • “She’s just so in the weeds,” Weinberg added. “She really knows what’s happening at all these companies from the ground level.”
    Sarah Guo's Wager - Colossus
  • What Harvey sells to law firms, underneath the software, is trust, and trust is delivered by humans. So the AI company built to do the work of lawyers has hired 200 lawyers of its own, many of whom teach other lawyers, in person, how to let the software do their job. Guo told me this is what much of Conviction’s portfolio is finding: as the models keep getting better, the work of helping a human get value out of them keeps getting bigger.
    Sarah Guo's Wager - Colossus
  • So could Vernal, who lives down the peninsula too. On a good run, it is a 40-minute drive to work. “I tell Mike,” Guo said, “that we locked ourselves in jail. But when we decided that AI was going to be like the Industrial Revolution, you don’t observe that from afar.” The whole setup is built to attract a particular kind of person: the worker ant leaving the biggest colonies behind.
    Sarah Guo's Wager - Colossus
  • A few blocks west is the Pioneer Building, OpenAI’s first headquarters; Elon Musk holds the lease, and after OpenAI moved out he installed xAI in its place. A short walk north is Mira Murati’s Thinking Machines, the lab she founded after leaving OpenAI as chief technology officer. Conviction backed it. The blocks between are thick with young AI companies. Someone began calling the neighborhood the arena, and the name stuck.
    Sarah Guo's Wager - Colossus
  • “If you hire eight people, or you raise more than $20 million,” Guo said, “it’s time.” Most tenants get the message. Alex Graveley, the first architect of GitHub Copilot, camped here while he built his company. Niki Parmar and Ashish Vaswani, two of the authors of the paper that introduced the Transformer, and with it more or less the present era, took a row of eight desks. Karpathy worked from these desks before he moved his family south, at which point he migrated to Conviction’s new outpost in Palo Alto.
    Sarah Guo's Wager - Colossus
  • “If you’re going to do something else,” Guo told her, “just make it the most aggressive thing that you could possibly be doing. I’m happy to call Kevin and we’ll find you a better job. But that [job] is not aggressive enough for you.” Kevin Weil was then OpenAI’s chief product officer.
    Sarah Guo's Wager - Colossus
  • Andrej Karpathy, before he joined Anthropic in May, worked out of Conviction’s office. Guo has been close to Jensen Huang, the founder of Nvidia, for more than a decade.
    Sarah Guo's Wager - Colossus
  • “There’s just a manic energy in Silicon Valley right now,” Elad Gil, one of the Valley’s most prominent investors, told me. “It’s been a really big shift in the last six months.”
    Sarah Guo's Wager - Colossus
  • He has not written a line of code since December. He also, like Friedman, has an agent in charge of his house. It’s called Dobby.
    Sarah Guo's Wager - Colossus
  • Promote the transformation of research methods in philosophy and social sciences towards a human-machine collaborative model, explore the establishment of new organizational forms for philosophy and social science research adapted to the era of artificial intelligence, and broaden research horizons and perspectives.
    Opinions of the State Council on Deepening the Implementation of the "Artificial Intelligence+" Action Plan (Central Cyberspace Administration Office)
  • Expedite the exploration of new AI-driven research paradigms and accelerate the process of major scientific discoveries "from 0 to 1". Accelerate the construction and application of large-scale scientific models, promote the intelligent upgrading of basic research platforms and major scientific and technological infrastructure, create open and shared high-quality scientific datasets, and improve the processing level of complex cross-modal scientific data.
    Opinions of the State Council on Deepening the Implementation of the "Artificial Intelligence+" Action Plan (Central Cyberspace Administration Office)
  • After graduation, Liang Wenfeng neither sought employment nor started his own business. He remained a retail investor, dabbling in the A-share market, constantly experimenting with quantitative and automated strategies to test their effectiveness. It's said that he once left Hangzhou, rented a room in Chengdu, and spent three years trading stocks behind closed doors.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • The moat formed by closed-source projects is short-lived. Publishing papers through open source doesn't mean losing anything; for technical personnel, being followed is very fulfilling. Therefore, open source is more like a cultural phenomenon.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • Liang Wenfeng has created his own success in his own way. He didn't follow the traditional expectations of a good student in university, nor did he study worldly social skills. He is a typical example of "Be Yourself" among Chinese university students, and also an example of contemporary educated youth changing their own destiny (and even the destiny of the nation) through entrepreneurship.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • "Old rascal, you are truly remarkable. I, Huang Yaoshi, am indifferent to 'fame,' and Master Yideng sees 'fame' as an illusion, but you, with an empty heart, have never harbored a single thought of 'fame,' yet you surpass us all."
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • Therefore, despite the overwhelming buzz surrounding his in-depth exploration project, he didn't publish an article, say a word, or record a video. As his university classmates, we weren't surprised by his low-key demeanor at all. Most people don't possess such composure and steadiness.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • Liang Wenfeng has always been low-key,
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • While none of the three were among the top students in their major, their competition results were outstanding.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • During his university years, Liang Wenfeng cycled around several provinces in East China, often sleeping on the ground in the wild at night, and the trip didn't cost him much money.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • He rarely attended classes, relying mostly on self-study. The author speculates this is because he felt the teachers' pace was slow and a waste of time, preferring self-study. The downside was that he didn't follow the teachers' highlighted points, which put him at a disadvantage during exams.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • This design seemed incredibly impressive at the time; we all exclaimed "Wow!" He humbly said the guitar's tone wasn't good, and it would be even better if it could automatically tune.
    Who is Liang Wenfeng, the founder of DeepSeek? - Zhihu
  • the combined peak node occupancy for V3 and R1 inference services reached 278, with an average occupancy of 226.75 nodes (each node contains 8 H800 GPUs). Assuming the leasing cost of one H800 GPU is $2 per hour, the total daily cost amounts to $87,072.
    open-infra-index/202502OpenSourceWeek/day_6_one_more_thing_deepseekV3R1_inference_system_overview.md at main · deepseek-ai/open-infra-index
  • As one of our investors, Magic Cube has a sufficient R&D budget, and also has an annual donation budget of several hundred million yuan, which was previously given to public welfare organizations. If necessary, some adjustments can be made.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • Driven by curiosity.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • Our goal is also very clear: not to focus on specific categories or applications, but to conduct research and exploration.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • An interesting detail is that in the early years, a friend who was equally eccentric and making "unreliable" drones in a Shenzhen urban village recruited him to join the company. Later, this friend built a company worth hundreds of billions of dollars called DJI.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • After graduation, instead of becoming a programmer at a big company like those around him, he hid in a cheap rental room in Chengdu, constantly experiencing setbacks in various scenarios. Eventually, he entered one of the most complex scenarios, finance, and founded Magic Square.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • The secret to Magic Square's growth is attributed internally to "selecting a group of inexperienced but high-potential individuals, as well as having an organizational structure and corporate culture that allows innovation to occur." They believe this is also the secret to how large model startups can compete with large companies.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • The founder of a leading quantitative fund believes that over the years, Magic Square has never followed any conventional path, but rather "followed its own way." Even if it is somewhat unconventional or controversial, "it dares to speak out frankly and then do things according to its own ideas."
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • The only exception is Magic Square: it started entirely with a local team and grew up on its own.
    The Magic Cube: The Path to a Large Model for an Invisible AI Giant
  • Gross, the person said, is a “slightly reluctant founder.”
    The Superintelligence of AI Investor Daniel Gross — The Information
  • And in the early days of Stripe, which was founded in 2010, Collison got some friendly programming assistance from Gross after the startup struggled to integrate its technology with a popular web framework, even though Gross wasn’t an investor in Stripe at the time. “He’s like, ‘Screw it. I’ll just do it myself.’ And then the same day, he just wrote [the integration] himself,” Collison said.
    The Superintelligence of AI Investor Daniel Gross — The Information
  • Hardcore innovation will only increase in the future. It’s not widely understood now because society as a whole needs to learn from reality. When this society starts celebrating the success of deep-tech innovators, collective perceptions will change. We just need more real-world examples and time to allow that process to unfold.
    Interview with Deepseek Founder: We're Done Following. It's Time to Lead.
  • Wang Huiwen took on all the losses himself, allowing others to exit unscathed. He made a decision that was most unfavorable to himself but beneficial to everyone else. I truly admire his integrity.
    Interview with Deepseek Founder: We're Done Following. It's Time to Lead.
  • Because we are tackling the hardest problems. The most attractive thing for top-tier talent is the opportunity to solve the world’s toughest challenges. In fact, top talent in China is often underestimated because hardcore innovation is rare, which means they rarely get recognized. We offer what they crave.
    Interview with Deepseek Founder: We're Done Following. It's Time to Lead.
  • I believe innovation is, first and foremost, a matter of belief. Why is Silicon Valley so innovative? Because they dare to try. When ChatGPT debuted, China lacked confidence in frontier research. From investors to major tech firms, many felt the gap was too wide and focused instead on applications. But innovation requires confidence, and young people tend to have more of it.
    Interview with Deepseek Founder: We're Done Following. It's Time to Lead.
  • Our hiring standards have always been based on passion and curiosity. Many of our team members have unique and interesting backgrounds. Their hunger for research far outweighs monetary concerns.
    Interview with Deepseek Founder: We're Done Following. It's Time to Lead.