Introducing Stitch AI: put agents to work, native inside the Stitch operating system.
Stitch AI fetches data, runs tasks and changes configuration across the platform, under the bank’s own permissions and regulatory constraints.
Claude Code tip: a verification gate your agent can't talk its way past, orchestrated by Fable 5.2.
Hand this to Claude Code to set it up ↓
the mistake almost everyone makes is letting the agent decide when the work is done. it will quietly edit its own tests to go green and call it shipped.
the fix is a gate. Fable 5.2 runs the loop, but nothing merges until a hook runs the real tests and a verifier that never wrote the code signs off.
Fable 5.2 (orchestrator, runs the loop)
├─ Opus 5.5 high → plan and build the change
├─ hook → run the real tests, reject on fail or edited tests
├─ Opus 5.5 medium → verifier, reads the diff cold before merge
└─ breaker → 3 failed loops → git reset to the last clean commit
every pass has to clear the hook and the verifier. the agent can't mark its own homework, and a broken run rolls back instead of piling fixes on fixes.
paste this into Claude Code and it wires the gate for you:
"Set up a verification gate for this repo.
Fable 5.2 orchestrates a plan, build, test, verify loop.
Opus 5.5 (high) plans and builds the change.
Add a pre-commit hook that runs the full test suite and rejects the commit if tests fail or any test file was changed.
Opus 5.5 (medium) runs as a verifier that reads the diff cold and must approve before merge.
If the same error repeats 3 times, reset to the last clean commit and start over.
Show me the gate before you run it."
one chat window, a gate nothing gets past.
This is f*cking insane. Microsoft CEO Satya Nadella:
"Business applications, that's probably where they'll all collapse in the agent era. They are essentially databases with a bunch of business logic
The business logic is all going to these agents. And once the AI is where all the logic is, people will start replacing the backends"
In just 2 minutes, he lays out why the software your company pays for every month is about to become a storage box the agents write into
Apps → Copilots → Agents → Software That Runs Itself
An app makes you click through its screens and follow its rules
An agent skips the screens, updates every system at once and keeps all the logic for itself
Anyone can learn one more app, almost nobody is learning to run the agents that replace it
Save it and watch it today
then read the article below ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
This is f*cking insane. Anthropic CEO Dario Amodei just went on the record about whose job goes first:
"Entry-level jobs in finance, consulting and tech are going to be first augmented, but before long replaced by AI systems. Somewhere between one and five years we'll see big effects
There will always be people who are using AI. Learn it now, or watch your role dry up"
In just 4 minutes, he lays out which jobs go first and who ends up ahead
Chatbots → Agents → Harness → AI Employees → Departments Run By AI
Surviving AI doesn't mean asking a chatbot to tidy up your emails. It means being the person who plugs agents into a company's work and keeps them running
Everything below is free. Start today:
Anthropic - Prompt Engineering Interactive Tutorial: learn to tell a model exactly what to do
github.com/anthropics/prompt…
Anthropic Courses: your path from first prompt to running agents
github.com/anthropics/course…
Anthropic Cookbook: copy-ready recipes for real work
github.com/anthropics/anthro…
Microsoft - Generative AI for Beginners: 21 lessons, no degree needed
github.com/microsoft/generat…
Microsoft - AI Agents for Beginners: build the agents that do the entry-level work
github.com/microsoft/ai-agen…
Hugging Face - Agents Course: free, certified, end to end
github.com/huggingface/agent…
Google Cloud - Generative AI: Google's own hands-on notebooks
github.com/GoogleCloudPlatfo…
OpenAI Cookbook: code for every common task
github.com/openai/openai-coo…DAIR.AI - Prompt Engineering Guide: the guide everyone keeps open
github.com/dair-ai/Prompt-En…
Andrej Karpathy - Neural Networks Zero to Hero: see what is under the hood
github.com/karpathy/nn-zero-…
CrewAI: run a small team of AI workers
github.com/crewAIInc/crewAI
Microsoft - AutoGen: put several agents on one job
github.com/microsoft/autogen
Awesome LLM Apps: hundreds of real builds to learn from
github.com/Shubhamsaboo/awes…DeepLearning.AI - Short Courses: free 1-hour courses from the people who teach this
deeplearning.ai/short-course…
Total cost: $0
You don't need a degree to stay ahead of this. You need to learn to run the agents doing the work
95% of people will watch AI eat their industry. The 5% who build with it will be the ones companies fight over
Save this and send it to someone whose job is on that list
then read the article below ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
This is f*cking insane. Anthropic CEO Dario Amodei just went on the record about whose job goes first:
"Entry-level jobs in finance, consulting and tech are going to be first augmented, but before long replaced by AI systems. Somewhere between one and five years we'll see big effects
There will always be people who are using AI. Learn it now, or watch your role dry up"
In just 4 minutes, he lays out which jobs go first and who ends up ahead
Chatbots → Agents → Harness → AI Employees → Departments Run By AI
Surviving AI doesn't mean asking a chatbot to tidy up your emails. It means being the person who plugs agents into a company's work and keeps them running
Everything below is free. Start today:
Anthropic - Prompt Engineering Interactive Tutorial: learn to tell a model exactly what to do
github.com/anthropics/prompt…
Anthropic Courses: your path from first prompt to running agents
github.com/anthropics/course…
Anthropic Cookbook: copy-ready recipes for real work
github.com/anthropics/anthro…
Microsoft - Generative AI for Beginners: 21 lessons, no degree needed
github.com/microsoft/generat…
Microsoft - AI Agents for Beginners: build the agents that do the entry-level work
github.com/microsoft/ai-agen…
Hugging Face - Agents Course: free, certified, end to end
github.com/huggingface/agent…
Google Cloud - Generative AI: Google's own hands-on notebooks
github.com/GoogleCloudPlatfo…
OpenAI Cookbook: code for every common task
github.com/openai/openai-coo…DAIR.AI - Prompt Engineering Guide: the guide everyone keeps open
github.com/dair-ai/Prompt-En…
Andrej Karpathy - Neural Networks Zero to Hero: see what is under the hood
github.com/karpathy/nn-zero-…
CrewAI: run a small team of AI workers
github.com/crewAIInc/crewAI
Microsoft - AutoGen: put several agents on one job
github.com/microsoft/autogen
Awesome LLM Apps: hundreds of real builds to learn from
github.com/Shubhamsaboo/awes…DeepLearning.AI - Short Courses: free 1-hour courses from the people who teach this
deeplearning.ai/short-course…
Total cost: $0
You don't need a degree to stay ahead of this. You need to learn to run the agents doing the work
95% of people will watch AI eat their industry. The 5% who build with it will be the ones companies fight over
Save this and send it to someone whose job is on that list
then read the article below ↓
This is f*cking insane. Anthropic CEO Dario Amodei was asked what a kid finishing high school should do now. His answer:
"AI can do entry-level law work. It can do entry-level finance. It can do entry-level consulting
We can't deny that it will also eliminate jobs, and probably a large number of them"
In a few minutes he names which jobs go first and the one skill that keeps you in the game
His own prediction: 50 percent of entry-level white-collar jobs disrupted in the next 1 to 5 years
Prompts → Agents → AI Coworkers → Entry-Level Work Done By AI
Farming to factories took generations. This time AI hits every step of a career at once
His advice fits in one line: "Of course, learn to use AI"
The people who start this year won't be competing with AI, they'll be the ones running it
Everything below is free. Start today:
Anthropic - Prompt Engineering Interactive Tutorial: the fastest way to learn to direct a model
github.com/anthropics/prompt…
Anthropic Courses: from your first prompt to running agents
github.com/anthropics/course…
Anthropic Cookbook: real task recipes you can copy today
github.com/anthropics/anthro…
Microsoft - Generative AI for Beginners: 21 lessons, no degree required
github.com/microsoft/generat…
Microsoft - AI Agents for Beginners: build agents that do the entry-level work for you
github.com/microsoft/ai-agen…
Hugging Face - Agents Course: free and certified, start to finish
github.com/huggingface/agent…
Google Cloud - Generative AI: Google's own hands-on notebooks
github.com/GoogleCloudPlatfo…
OpenAI Cookbook: code for every common task
github.com/openai/openai-coo…DAIR.AI - Prompt Engineering Guide: the reference everyone links to
github.com/dair-ai/Prompt-En…
Andrej Karpathy - Neural Networks Zero to Hero: understand what is under the hood
github.com/karpathy/nn-zero-…
CrewAI: run a small team of AI workers
github.com/crewAIInc/crewAI
Microsoft - AutoGen: put several agents on one job
github.com/microsoft/autogen
Awesome LLM Apps: hundreds of real builds to learn from
github.com/Shubhamsaboo/awes…DeepLearning.AI - Short Courses: free 1-hour courses from the people who teach this
deeplearning.ai/short-course…
Total cost: $0
You don't need another year of tuition to learn this. You need to start telling AI what to do
95% will graduate and compete with AI for the same entry-level job
The 5% who learn to run it this year will manage the agents doing that job
Save this and send it to someone graduating this year
Then read the full Graph Engineering guide below ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
This is f*cking insane. Former Google CEO Eric Schmidt just told everyone the easiest way to get rich in the next two years, and it's not stocks or crypto:
"If you really want to make money, it's actually easy, found an agentic AI company, I mean build an agent to do something
If you build a better agent than anyone else, you can probably build one incredible company and own it yourself"
The man who ran Google for 10 years lays out in under 30 seconds why the next year or two is the agentic period, and why the window won't stay open
Prompts → Claude Code → Loops → Graphs → Gates → A Company You Own Yourself
Save this before it slips off your feed
Everything below is free:
Building effective agents (Anthropic)
anthropic.com/engineering/bu…
How Anthropic built its multi-agent research system
anthropic.com/engineering/mu…
Harnesses for long-running agents
anthropic.com/engineering/ef…
Writing tools for agents
anthropic.com/engineering/wr…
Claude Code 101
anthropic.skilljar.com/claud…
Claude Code in action
anthropic.skilljar.com/claud…
Prompt engineering
platform.claude.com/docs/en/…
Subagents (parallel work)
code.claude.com/docs/en/sub-…
Hooks (checks that can say no)
code.claude.com/docs/en/hook…
CLAUDE.md + memory
code.claude.com/docs/en/clau…
Skills
code.claude.com/docs/en/skil…
MCP
code.claude.com/docs/en/mcp
LangGraph (graphs of agents)
langchain-ai.github.io/langg…
Karpathy's autoresearch (an agent in a loop)
github.com/karpathy/autorese…
All of this costs $0
95% will spend the agentic period scrolling AI threads. The 5% who ship agents will own the companies that come out of it
Bookmark it and open the first link tonight
Then read the full Graph Engineering guide below ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
Your agent deleted a folder and you have 10,000 lines of logs. XERJ Reader finds why in under a second.
Ask "why did the agent delete my folder?" over the logs -> it highlights the exact tool call that did it.
It does not summarize or guess. It points at the line and its position in the file.
An LLM agent burns a minute and real tokens. Reader costs $0.01 per 1M input, output free.
A minute of agent time, or one cent.
I've built a new AI model that is not allowed to generate text. Instead, it produces a verifiable program that answers your prompt using the original reference.
XERJ Reader-1 is a neurosymbolic model for working with data. It points to the exact words in your document, writes a small program over them, and exact code computes the answer:
• 20-90x faster
• 10-1,000x cheaper (output tokens free)
• every answer is a program you can replay
The shortest path to automating data work.
This is f*cking insane. Google CEO Sundar Pichai says today's AI will look like a flip phone in three years:
"In three years, today's AI will look as primitive as a flip phone. The progress we're about to see will surprise people"
In under 2 minutes, he explains why this moment is different from every other tech cycle
He calls it rare: the curve is so steep that in three years people will ask "you mean you couldn't do those things?"
His bet on where it goes is one word: agents. Everyone is laying the foundation now, but soon they'll be everywhere
People won't sit and prompt. They'll lean on agents to get things done in a seamless way
And any content will flow from any modality to any modality. He points at Nano Banana, Veo and Omni as the first hints
Chatbots → Prompts → Agents → Harness → Loops → It Just Gets Done
The people who win the next three years aren't the ones watching the curve. They're the ones building on it
Bookmark this before your feed buries it
Every course and repo below costs nothing:
A free AI course for non-programmers
elementsofai.com
AI Python for beginners, taught by Andrew Ng
learn.deeplearning.ai/course…
Interactive prompt engineering tutorial
github.com/anthropics/prompt…
Gemini in your terminal
github.com/google-gemini/gem…
The OpenAI cookbook
github.com/openai/openai-coo…
Build resilient agents as graphs
github.com/langchain-ai/lang…
Agents that work as a team, each with a role
github.com/crewAIInc/crewAI
An agent that researches any topic in depth
github.com/assafelovic/gpt-r…
An agent that clicks through sites and fills forms
github.com/browser-use/brows…
A memory layer so your agents remember context
github.com/mem0ai/mem0
Automate busywork with drag-and-drop workflows
github.com/n8n-io/n8n
18 free lessons on building agents
github.com/microsoft/ai-agen…
Total cost: $0
Everyone will nod at the flip-phone line and scroll past. Almost nobody will start building for the version that's coming
95% will wait three years to see if he's right, the 5% who build now will own the head start
Save it and send it to anyone still treating AI like a toy
Then use the guide below to build your harness before the curve leaves you behind ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
This is f*cking insane. Warren Buffett was asked what to do when a stock drops 50%. His answer had nothing to do with the stock:
"If you buy stocks, you must be mentally prepared for them to fall 50% or even more"
In 2 minutes, he takes apart the one thing that actually blows up investors, and it isn't the market
Berkshire's own stock has fallen roughly 50% three separate times. Nothing was wrong with the business. Prices and business values just don't move together
That gap is where people destroy themselves. His one hard rule: never buy with borrowed money, because when it drops you won't survive the pressure
Nobody can call the bottom. Nobody. He buys anyway, because he's looking 20 to 30 years out
Panic → Rules → A Checklist → A System → The Investor Who Holds
The hardest part isn't buying the bottom or selling the top. It's staying rational while the screen is red and knowing if the business is still worth believing in
So take yourself out of it. Let a system hold the discipline you can't hold at 3am
Bookmark this before your feed buries it
Every tool below costs nothing:
OpenBB, a free research terminal for any stock
github.com/OpenBB-finance/Op…
Microsoft's AI quant research platform
github.com/microsoft/qlib
Backtest a strategy before you risk a cent
github.com/QuantConnect/Lean
A rules-based bot that trades without emotion
github.com/freqtrade/freqtra…
Technical indicators for your own screens
github.com/twopirllc/pandas-…
Pull price data from any exchange
github.com/ccxt/ccxt
Event-driven backtesting in Python
github.com/mementum/backtrad…
An agent that researches a company in depth
github.com/assafelovic/gpt-r…
Anthropic's SDK for building your own agents
github.com/anthropics/claude…
A memory layer so your agents remember your rules
github.com/mem0ai/mem0
Automate your research with drag-and-drop workflows
github.com/n8n-io/n8n
18 free lessons on building agents
github.com/microsoft/ai-agen…
Total cost: $0
A crash doesn't test your picks. It tests your mindset
95% will panic-sell the next red week, the 5% who let a system decide will still be holding in 20 years
Save it and send it to anyone watching their portfolio bleed
Then use the guide below to build the system that stays rational when you can't ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a
This is f*cking insane. Microsoft CEO Satya Nadella:
"Business applications, that's probably where they'll all collapse in the agent era. They are essentially databases with a bunch of business logic
The business logic is all going to these agents. And once the AI is where all the logic is, people will start replacing the backends"
In just 2 minutes, he lays out why the software your company pays for every month is about to become a storage box the agents write into
Apps → Copilots → Agents → Software That Runs Itself
An app makes you click through its screens and follow its rules
An agent skips the screens, updates every system at once and keeps all the logic for itself
Anybody can learn one more app, almost nobody is learning to run the agents that replace it
Save and watch it today
Then read the full Graph Engineering guide below ↓
Hanako @hanakoxbt Aug 23 Article Loops and Graphs: how to stop babysitting agents and only approve the last step (full course) You check every step your agents take. Not because you want to, but because nothing else does. Getting out of that chair takes two things, and almost everyone builds only one of them. A loop makes a