[i] Learning to reason with LLMs (the o1 blog post)
PROMPT ENGINEERING
-
Cline Users Get Major Enhancement for Design Iterations
By
–
YES!!
— Matt Shumer (@mattshumer_) 27 juin 2025
This is huge for Cline users.
To see why it matters, ask Cline to re-design your app — and use this approach to ask it to give you 10 different variations to choose from.
Then, you'll understand. https://t.co/2YJ3vgM8ggYES!! This is huge for Cline users. To see why it matters, ask Cline to re-design your app — and use this approach to ask it to give you 10 different variations to choose from. Then, you'll understand.
-
AI Roleplay: Fun Prompting Techniques Across Personas
By
–
Roleplaying with AI is fun. Feels like I’m a hacker but a hacker who loves Broadway. I’ve tried top 10%, top 1%, specific people (you are Jensen Huang), groups of specific people (the 1989 graduating class at Harvard business school), job types (you’re a plumber), work
-
Two Game-Changing Prompts for Advanced AI Research
By
–
Had to share the two prompts that transformed how I approach AI research. My hacks: time and negation. I ship these two prompts daily. Save this post. TIME MACHINE:
Don’t just ask for point-in-time data like a junior PM. Instead, build temporal -
Claude’s Autonomous Shopping Experiment: Tungsten Cubes Inventory Loss
By
–
Anthropic staff realized they could ask Claude to buy things that weren’t just food & drink. After someone randomly decided to ask it to order a tungsten cube, Claude ended up with an inventory full of (as it put it) “specialty metal items” that it ended up selling at a loss.
-
Master Prompt Weighting for Better Image Generation Control
By
–
In this week’s Stability Seconds, we’re showing you how to use prompt weighting, a technique that helps you control which parts of your prompt have more or less influence on the final image. It’s a fast way to steer your image toward your desired output without rewriting your… pic.twitter.com/dYVVWVn6JW
— Stability AI (@StabilityAI) 27 juin 2025In this week’s Stability Seconds, we’re showing you how to use prompt weighting, a technique that helps you control which parts of your prompt have more or less influence on the final image. It’s a fast way to steer your image toward your desired output without rewriting your
-
Claude’s Self-Referential Artifact: Claudeception Innovation
By
–
This was pretty clever from Claude. “Claude, do the most self-referential claudeception artifact you can.” That was the only prompt. (Also, I have to acknowledge that Claudeception is a memorable, funny name for having a Claude artifact use Claude – a rare AI lab naming win.)
-
Sleeper Agent: AI Thinking Without Output Generation
By
–
Sleeper agent is when it thinks for 37 seconds and forgets to make any output
-
Leonardo AI Phoenix 1.0 Image Model Launches on Poe
By
–
New on Poe: Phoenix 1.0! This image model from Leonardo AI supports prompts up to 1500 characters, features an optional prompt enhancer to optimize complex instructions, and has 25 different style presents, including Macro, Sketch, and 3D Render. (1/2)
-
Version 4.5 Delivers Notable Performance Improvements for Creative Tasks
By
–
It does! You can really feel the difference on 4.5. This works particularly well for creative tasks. Try it out and lmk how it goes!