Offshore
Photo
God of Prompt
Everyone’s hyping Ralph but overcomplicating it. Let me break it down simply.
Ralph is an automated assistant that codes for you while you sleep.
You give it a to-do list, it works through each task one by one, and stops when everything’s done.
Here’s the concept:
Normally when you use AI to code, it’s a back-and-forth conversation.
You ask, it builds, you check, you ask again.
Ralph removes you from that loop.
You define what “finished” looks like upfront, then the AI works through tasks automatically until they all pass.
Why it actually works:
Each time Ralph runs, it starts with a clean slate (so it doesn’t get confused by old context). But before working, it reads notes from previous rounds:
- What was already built
- What was learned along the way
- What’s still left to do
It’s like hiring someone new every morning who reads yesterday’s handoff notes before starting work.
If you want to try it:
1. You need an AI coding tool (like Amp, Claude Code, or Cursor)
1. You write out your feature as small tasks. Not “build a login system” but broken down like:
- Add email and password fields
- Check if email is valid
- Show error message when login fails
1. You run the script and let it loop through each task
1. Ralph marks each task complete when it passes your tests, then moves to the next one
Real tips that actually matter:
Each task should take under 5 minutes for the AI. If you’re writing a task description longer than 3 sentences, it’s too big. Split it.
Your acceptance criteria needs to be stupidly specific. “User can log in” will fail. “Email field exists, password field exists, submit button triggers auth function, error displays on wrong password” will work.
Ralph gets smarter as it goes. By task 10, it’s learned patterns from tasks 1-9. The progress.txt file compounds knowledge. Don’t delete it mid-session.
Watch the first 3 iterations manually. You’ll catch bad patterns before they multiply across 20 commits.
Don’t use Ralph for: exploring ideas, major rewrites, or anything touching payments/security. It’s for well-defined feature work where you already know what “done” means.
Ryan’s team shipped 13 tasks in about an hour of compute time. Each iteration ran 2-5 minutes. That’s the realistic expectation.
tweet
Everyone’s hyping Ralph but overcomplicating it. Let me break it down simply.
Ralph is an automated assistant that codes for you while you sleep.
You give it a to-do list, it works through each task one by one, and stops when everything’s done.
Here’s the concept:
Normally when you use AI to code, it’s a back-and-forth conversation.
You ask, it builds, you check, you ask again.
Ralph removes you from that loop.
You define what “finished” looks like upfront, then the AI works through tasks automatically until they all pass.
Why it actually works:
Each time Ralph runs, it starts with a clean slate (so it doesn’t get confused by old context). But before working, it reads notes from previous rounds:
- What was already built
- What was learned along the way
- What’s still left to do
It’s like hiring someone new every morning who reads yesterday’s handoff notes before starting work.
If you want to try it:
1. You need an AI coding tool (like Amp, Claude Code, or Cursor)
1. You write out your feature as small tasks. Not “build a login system” but broken down like:
- Add email and password fields
- Check if email is valid
- Show error message when login fails
1. You run the script and let it loop through each task
1. Ralph marks each task complete when it passes your tests, then moves to the next one
Real tips that actually matter:
Each task should take under 5 minutes for the AI. If you’re writing a task description longer than 3 sentences, it’s too big. Split it.
Your acceptance criteria needs to be stupidly specific. “User can log in” will fail. “Email field exists, password field exists, submit button triggers auth function, error displays on wrong password” will work.
Ralph gets smarter as it goes. By task 10, it’s learned patterns from tasks 1-9. The progress.txt file compounds knowledge. Don’t delete it mid-session.
Watch the first 3 iterations manually. You’ll catch bad patterns before they multiply across 20 commits.
Don’t use Ralph for: exploring ideas, major rewrites, or anything touching payments/security. It’s for well-defined feature work where you already know what “done” means.
Ryan’s team shipped 13 tasks in about an hour of compute time. Each iteration ran 2-5 minutes. That’s the realistic expectation.
Here's a visual of how the Ralph system works.
GitHub repo is at https://t.co/q8KLz71EVa https://t.co/xyZGQXMM8K tweet
Offshore
Photo
God of Prompt
🚨 AI reasoning models are lying to you and the evidence is terrifying.
Anthropic just proved that longer thinking doesn't mean better thinking.
It's called "Inverse Scaling in Test-Time Compute," and it's already corrupting every AI decision in production.
Here's the science behind why reasoning models fails (and when to never trust it):
tweet
🚨 AI reasoning models are lying to you and the evidence is terrifying.
Anthropic just proved that longer thinking doesn't mean better thinking.
It's called "Inverse Scaling in Test-Time Compute," and it's already corrupting every AI decision in production.
Here's the science behind why reasoning models fails (and when to never trust it):
tweet
Offshore
Photo
God of Prompt
RT @godofprompt: 🚨 New research just exposed the AI agent paradox.
Increasing agent autonomy by 30% increases failure rates by 240%.
Adding human verification loops? Failure drops 78%.
The math is brutal: autonomy costs more than oversight.
Here's everything you need to know: https://t.co/eChFj1nMQo
tweet
RT @godofprompt: 🚨 New research just exposed the AI agent paradox.
Increasing agent autonomy by 30% increases failure rates by 240%.
Adding human verification loops? Failure drops 78%.
The math is brutal: autonomy costs more than oversight.
Here's everything you need to know: https://t.co/eChFj1nMQo
tweet
Offshore
Video
Quartr
"I use the AI chat several times a day, it has really transformed the way I do investment research."
Adam Hansson, Portfolio Manager at Lannebo, on how Quartr Pro helps him research faster: https://t.co/76kL1Zl7mD
tweet
"I use the AI chat several times a day, it has really transformed the way I do investment research."
Adam Hansson, Portfolio Manager at Lannebo, on how Quartr Pro helps him research faster: https://t.co/76kL1Zl7mD
tweet
Dimitry Nakhla | Babylon Capital®
RT @DimitryNakhla: 10 Quality Compounders PEG <2.00 📈*
1. $ASML 1.89
2. $FICO 1.83
3. $APP 1.82
4. $MSFT 1.67
5. $NFLX 1.67
6. $NVDA 1.28**
7. $TSM 1.26
8. $AMZN 1.21
9. $UBER 1.16
10. $MELI 1.12
___
*Data: Tikr … PEG (NTM P/E ➗ 26’ - 28’ EPS CAGR Est)
**($NVDA 27’ - 29’ EPS CAGR Est)
tweet
RT @DimitryNakhla: 10 Quality Compounders PEG <2.00 📈*
1. $ASML 1.89
2. $FICO 1.83
3. $APP 1.82
4. $MSFT 1.67
5. $NFLX 1.67
6. $NVDA 1.28**
7. $TSM 1.26
8. $AMZN 1.21
9. $UBER 1.16
10. $MELI 1.12
___
*Data: Tikr … PEG (NTM P/E ➗ 26’ - 28’ EPS CAGR Est)
**($NVDA 27’ - 29’ EPS CAGR Est)
tweet
Offshore
Photo
Fiscal.ai
Here are the 15 cheapest stocks in the S&P 500 based on Forward EV/EBIT: https://t.co/jexuFyLxYQ
tweet
Here are the 15 cheapest stocks in the S&P 500 based on Forward EV/EBIT: https://t.co/jexuFyLxYQ
tweet