Just posted a video on Instagram showing how I got Claude to track what US politicians and Wall Street whales are buying and mirror those trades automatically (paper trading, so zero real money at risk while you learn it). A bunch of you asked how it actually works, so I dropped the full written guide right here in the community. It walks through all 3 levels: 1. Setup — connect Claude to a free brokerage and place your first trade just by talking to it 2. Copy trading — a bot that finds the top-performing member of Congress and copies their positions on a schedule 3. Options income — an automated "wheel strategy" that collects premiums whether the stock goes up, down, or sideways The whole thing is no-code. You're literally just talking to Claude. Video - https://www.instagram.com/p/DZI1_0xgtQI/ 👉 https://flicker-celestite-7b6.notion.site/The-Claude-Trading-Bot-Guide-Copy-Congress-Automate-the-Market-374d180d8c808136a0c6ca44215caaf1?source=copy_link

This is one of the wilder AI stories of the week, and it's a perfect window into how fast the frontier is moving right now. Let me break down what's real vs. what's hype 👇 What's confirmed: A new Anthropic model identifier — claude-oceanus-v1-p — surfaced on June 3rd. It showed up selectable right inside the Claude Console Workbench. I caught it sitting in the model picker myself (screenshot below). Red teamers — the security folks who stress-test a model before release — reportedly got access that same day. Oceanus is understood to build on the Claude Mythos Preview foundation — the "too dangerous to release publicly" model that's been running in cybersecurity programs. So this looks like the next step in that Mythos line, not a random side project. Here's the part the hype threads are skipping: Within hours of hitting validated red teamers, someone allegedly resold API access to the unreleased model through a proxy service — at $16 per million tokens, way above normal enterprise rates. Anthropic reportedly paused access for the whole red-team group and opened an internal investigation. Read that again. A frontier model leaked, got resold on a grey market, and the security review got shut down — all in the same day. My take on why this actually matters: Everyone in the replies is screaming "next Claude is weeks away, let's go." Maybe. But that's the boring read. The real signal here is about control — and it's the exact thing I keep hammering on with every Mythos update. Anthropic's whole strategy with these high-capability models is gated, careful, locked-down access. Mythos wasn't released publicly specifically because it was too capable on the cybersecurity side. And here's a model in that same lineage getting leaked and resold before it even cleared red-teaming. That's the central tension of frontier AI in 2026 in one news cycle: the more capable the model, the more valuable it is to steal, and the harder it is to contain. What this means for you as a builder or operator:

This isn't a leak. It's not a rumor. Anthropic just published internal data on how much of Claude's development is now being done by… Claude. The numbers are nuts. Read these slow 👇 → 80%+ of all code merged into Anthropic's codebase is now written by Claude. (Before Claude Code launched in Feb 2025, that number was in the low single digits.) → Anthropic engineers now ship 8x as much code per quarter as they did from 2021–2025. → On a standard test — "take this small AI training script and make it faster" — a skilled human takes 4–8 hours to reach 4x speedup. Claude Opus 4 (May 2024) averaged ~3x. Mythos Preview (April 2026)? ~52x. → When shown a research session where a human made a wrong turn, Mythos Preview picked a better next step 64% of the time — up from 22% in 2024. → One April 2026 example: Claude shipped 800+ fixes that cut a class of API errors by 1,000x. Engineering estimate of how long that would've taken a human: four years. Anthropic's own framing: this is a possible path to recursive self-improvement — AI autonomously building a more capable successor. They say it's not here yet and not inevitable. But it's "happening faster than we thought." My read on what this actually means: The "AI builds AI" headline is the dramatic part. The part you should actually internalize is the rate of acceleration itself. Look at the task-length curve they shared. Opus 3 (Mar 2024) could reliably handle 4-minute tasks. Sonnet 3.7 (a year later) — 90 minutes. Opus 4.6 (a year after that) — 12 hours. Mythos Preview can hold a session for at least 16 hours and that's the top of what they can currently measure. Task length is doubling roughly every 4 months. That used to be every 7. The curve is bending. What this means for builders: Whatever you couldn't automate 6 months ago — you can now. The "the AI can't hold a long enough chain of thought to do this end-to-end" objection is dying in real time. Workflows that needed three orchestrated agents last quarter can be one agent in 90 days. Plan accordingly.

Most people use Claude Code completely wrong. The defaults teach you to type into one chat, hit enter, wait, type again. That's the loop. The guy who actually built Claude Code does basically the opposite. Boris Cherny runs ~15 Claude Code sessions at the same time, barely types most of his prompts, and dispatches work like a team lead. I read every public post he's made and put the whole thing on one page. The 7 moves: 1. Parallel sessions — 5 in terminal + 5–10 in browser. One session, one job. Opus + thinking mode for all of them 2. CLAUDE.md — the rules file Claude loads automatically. Every time it gets something wrong, you write the fix in. Compounding engineering — month 3 has 50 rules nobody else has 3. Plan mode — Shift+Tab+Tab. Pour energy into the plan, then auto-accept the implementation 4. Slash commands + sub-agents — .claude/commands/ + .claude/agents/. Boris uses /commit-push-pr, /simplify, /verify, /go dozens of times a day 5. /loop — local background loops up to 3 days. His 4: /babysit, /slack-feedback, /post-merge-sweeper, /pr-pruner 6. /schedule — same idea, runs in the cloud. Laptop can close. GitHub events can trigger it 7. Verification (the #1 tip) — "grill me on these changes", "prove it works", "scrap this and implement the elegant solution" The adoptable system if you're not a developer: 2–3 parallel sessions (not 15) + CLAUDE.md + plan-first discipline + named prompts for anything you do twice + 1 loop during workday + 1 schedule overnight + a verification line on every important task. If you do only ONE thing from the video: add a verification line. The quality jump is immediate. 📎 Full Boris Cherny guide PDF pinned below — every move with the setup steps + the Keep-vs-Skip cuts for non-devs. Comment which move you're copying first 👇 [Watch the video here ▶️]

Quick heads up for anyone on the fence about AI Accelerator+ — Once we hit 50 members, the price goes up. We're at 38 right now, so that's 12 spots left at founders pricing. After that, it's gone for good. If you've been thinking about joining, this is the window. Here's what you actually get inside: 💎 Weekly office hours — 1 technical session with Keshav + 1 Q&A with me 💎 Monthly breakdowns of the newest AI tools & releases (so you're never behind) 💎 Full training on vibe coding apps + building agentic workflows 💎 24/7 technical support when you get stuck 💎 Direct feedback from me and my team on what you're building 💎 Monthly competitions + guest speakers + job opportunities This is for entrepreneurs and business owners who want to use AI to automate work, save time, and scale — without needing to become technical. Whether you're just starting, running a business, or building an agency, you'll get the exact skills to build real AI workflows and turn them into income. 🔥 New unlock at 50 members: we're kicking off monthly workshop calls — live, hands-on builds where we work through real tools and workflows together. Everyone who joins before the cap gets these locked in at founders pricing. 12 spots. Then the price moves. 👉 https://www.skool.com/theaiaccelerator/about