Hayden So
3 followers · 3 following · 794 views
on the atlas — 19
- Red hat merchant1 savers
- mckinsey.com/~/media/mckinsey/featured insights/middle east and africa/the closest look yet at chinese economic engagement in africa/dance-of-the-lions-and-dragons.pdf#page=7.141 savers
- China’s Great Transformation - by Nicholas Mulder1 savers
- General Intelligence Company | A Day in the Life of the World’s Most AI Forward Company2 savers
- gog — Google Workspace in your terminal1 savers
- How we rebuilt Next.js with AI in one week2 savers
- Language Models will be Scaffolds | Alex L. Zhang2 savers
- Philipp Schmid on X: "Here is how @SpotifyEng uses coding background agents for thousands of code migrations and what they learned: - Define desired verifiable end states explicitly, not strict todo steps. - Code examples improved output reliability. - Agent has access to 3 tools, verify, git and https://t.co/SlXpCk3nPt" / X1 savers
- I'm joining a16z | Andreessen Horowitz1 savers
- I got fooled by AI-for-science hype—here's what it taught me1 savers
- AI safety technical research - Career review1 savers
- swarm intelligence phd - Google Search1 savers
- Investing for the AI transition – Educating Silicon1 savers
- sea glider bladder - Google Search1 savers
- Mini Blog Post 19: On systems - Living a life of zero willpower — Neel Nanda16 savers
- 2402.018011 savers
- Top Posts — Neel Nanda6 savers
- Dear Ben… What is MCP (Model Context Protocol)? | Ben's Bites1 savers
- Playing Chess with Peter Thiel1 savers
highlights — 9
As a result, China had the dual benefit of offering what the West needed at the right time, and of learning from the mistakes and failures of others. There was a good deal of chance in all this. An earlier development takeoff would have seen the PRC compete much more directly with the Asian Tigers, and perhaps liberalize its financial flows in such a way that it would have been exposed to a Japanese-style asset crash. And a greater embrace of debt, like the Eastern European economies in the 1970s, would likely have imperiled the CCP’s hold on power by the 1980s, as we know from the recent fina…
China’s Great Transformation - by Nicholas MulderDefine a task ("implement the next/navigation shim with usePathname, useSearchParams, useRouter"). Let the AI write the implementation and tests. Run the test suite. If tests pass, merge. If not, give the AI the error output and let it iterate. Repeat.
How we rebuilt Next.js with AI in one weekFor the second half of this decade, my intuition tells me to bet differently. I’m not saying that scaling is dead; scaling is quite literally the key to everything in a data-driven strategy like deep learning. It’s more that language models are really good now: so good, that I theorize that existing neural language models are actually severely underutilized. I am implying that they are much better at general task solving than what we naively use them for. We’ve spent the better half of this decade exhausting every axis of scale we can find (e.g. data, compute, model capacity) in hopes that the…
Language Models will be Scaffolds | Alex L. ZhangOne lesson hiding in plain sight here is that most of the audience of any successful post does not actually read it. They are told it by someone who did read it. There’s a primary audience who carefully reads the piece and does the cognitive work of “restructuring their consciousness” (see Walter Ong, Orality and Literacy) around good writing. And then there’s a secondary audience, who are re-told the content, either verbally (including group chats, podcasts, Youtube) or in other oral formats like Twitter. (This is broadly true both inside and outside of organizations; e.g. a lot of work goes …
I'm joining a16z | Andreessen HorowitzMost importantly, we think PhDs shouldn’t be thought of as a pure delay to your impact. You can do useful work in a PhD, and generally, the first couple of years in any career path will involve a lot of learning the basics and getting up to speed. So if you have a good mentor, work environment, and choice of topic, your PhD work could be as good as, or possibly better than, the work you’d do if you went to work elsewhere early in your career. And if you suddenly receive evidence that we have less time than you thought, it’s relatively easy to drop out.
AI safety technical research - Career reviewSocial I have a spreadsheet keeping track of when I last spoke to friends, and reminding me when to next check in/schedule a call This was such a good idea - it’s really easy to forget people exist when I’m not seeing them in real life, and this rectifies that without consuming mental energy With a robust system, this also ensures I keep in touch with less close friends, who I eg want to check in on once every few months
Mini Blog Post 19: On systems - Living a life of zero willpower — Neel NandaProductivity On systems - building routines and systems into your life, so that you do the right thing by default rather than by “trying harder” Procrastination On procrastination - my toolkit for overcoming procrastination, and taking actions that achieve my values
Top Posts — Neel NandaSo, can companies support MCP by essentially wrapping their existing APIs inside an "MCP Server"? And where would that server actually live – on my machine, or theirs?
Dear Ben… What is MCP (Model Context Protocol)? | Ben's BitesThe game finishes up Thiel asks who wants to play and I immediately leap forward to challenge him. I quickly move to reset the board before anyone can suggest something more egalitarian.
Playing Chess with Peter Thiel