Movez

@RohOnChain They are building AI for decades already 😅

@0xMovez exactly movez, many people still dont realize the top quant firms are building proprietary AI infra and not just prompting chatgpt or claude

Jane Street pays $750k/year for engineers who build AI trading systems from scratch. This 15-minute talk by Jane Street senior engineer shows the exact system behind a $750k/year skillset. here's what he covers: • why generic AI fails at a top trading firm • custom models trained on their own trading code • real-time grading of every AI-written change • one AI engine across the whole trading desk most retail traders are still using generic AI tools - while the top hedge funds train their own models Watch this video then read article below

@RuujSs Yeah this is true !

@0xMovez Jane Street pays $750k/year for a reason. And this is it!

@0xCodez Amazing share Codez ! Keep delivering alpha

Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. This 1-hour Stanford lecture on "LLM architecture" gives you exact pipeline they get paid $750K/year for. Attention variants + normalization + positional encoding - the real stack. Bookmark it & watch today. Then read the article below.

Anthropic pays $750,000+ a year for engineers who can build LLM architectures from scratch. This 2-hour Stanford lecture gives you the exact pipeline LLM engineers get paid $750K/year for. Data + architecture + scaling laws + post-training. Bookmark it & watch today. Then read article below.

@Just_Codly What surprised me is that the key thing is data > architecture.

@0xMovez what part of the pipeline actually surprises people the most when they see how it works?

@nagarjuncreates For sure, you need good skills to build it, but once you have a pipeline of what to learn, it just becomes a plan for you.

@0xMovez That’s pretty cool. Makes you realize how much skill goes into these models. I might check out that lecture tonight.

@RohOnChain Wow. I was searching how hedge funds use AI on weekends, bro, and you just dropped an article. Going to read it now. Thanks for the share, bro.

@rewind02 damn, another banger watch by YC, going to deep dive right now. Thanks for share brother

A YC partner just described a company that rewrites its own code overnight without a single human involved this isn't a concept. it's already running at YC right now the talk is 13 minutes. it will change how you think about what a company even is here's what most founders are still missing: > every company today is organized like a Roman legion - humans passing information up and down a hierarchy > "20% more productive with AI co-pilots" that's a better engine on an old car > YC built an agent that watches every query their team makes > the same loop works for product: agent finds funnel friction, runs an A/B test, picks the winner, deploys > "if it didn't get recorded, it didn't happen to your AI." every Slack, email, and call you don't capture - you're throwing away your company brain > YC fed 2,000 hours of office hour recordings into AI over a weekend. output: a 150-page user manual. more accurate than the one built over 10 years what this means for how you build right now: > stop measuring headcount. start measuring token consumption. companies at YC demo day are hitting 5x more revenue per employee than 18 months ago > record everything. meetings, calls, decisions, Slack threads. if it isn't captured, your AI can't learn from it > treat your software as disposable. your data and domain knowledge are the asset. regenerate the software whenever models improve > eliminate middle management from your org chart. replace coordination layers with AI loops and directly responsible individuals > ask yourself: which parts of my company can run this loop without a human in the middle? the founders who build companies as recursive self-improving AI systems in the next 12 months will not just be more efficient they will be a fundamentally different type of organization than everyone else FULL TALK BELOW 13 minutes. the most important reframe on company building I've seen this year

@rewind02 Yeah. But its a reality today !

@0xMovez $750k skillset for free is wild

@AnatoliKopadze Legendary watch, brother. Watching each interview with Boris.

Boris Cherny: "Claude Code is 100% written by Claude Code, Cowork is 100% written by Claude Code" across all of Anthropic, 90% of code is written by Claude lawyers, designers, finance people are building real things with Claude Code right now with no CS degree, no background, nothing the people who built the tool use the tool to build the tool think about what that means for everyone else the guide on everything Claude can actually do is in the article below

@monokern havent seen him on Anthropic stages before

@0xMovez does the person who is giving the lecture work for the company Anthropic?

@marfinxx If you learn how to do it, I’m sure it’s possible for any one of my followers.

@0xMovez all the money in the world is here now

@RuujSs Wow, that sounds really interesting, bro. What was the result? Maybe drop some article on this topic

@0xMovez No haha, mostly learning. Went deep into understanding markets, decision making, and financial systems.

@RuujSs Damn. Real legend. What it was for ? Trading ?

@0xMovez Yep I did like 15 months ago…

@EagleVisions $750k a year is a good money for been an engineer bro

@0xMovez I'm going to become an engineer for build llm architecture

@Nikitont Share your insights after you read it, bro!

@0xMovez I haven't done it yet, I'll keep this article for myself.

@Nikitont it time to start know, until it too late mate. Btw have you tried to build your own LLM before ?

@0xMovez I know what I'm going to spend 2 hours on.

@dunik_7 amazing watch Dunik ! Booked it

Demis Hassabis is racing to build AGI and he keeps admitting the smartest AI on earth still forgets everything the moment you close the tab. Nobel laureate. runs DeepMind. on stage at Google I/O he says we're "on the cusp of a new golden age of discovery" then names the thing blocking it: AI can't carry what it learns from one session to the next. the model that aced your problem yesterday wakes up today having never met you. DeepMind needs a breakthrough to fix that at the frontier. you need a text file and twenty minutes: / a file your agent reads at the start and rewrites at the end / one filter for what survives "would this change how it acts next time?" / a weekly job that consolidates its memory while you sleep the labs are solving memory for AGI. you only have to solve it for Tuesday

@RuujSs have you tried to build your LLM before brother ? For trading for example

@0xMovez This lecture on LLM architecture + the article attached is solid combo!