Sometimes when I demo AI, I show it turning cover letters into goofy formats (poetry, etc) as an introduction to the idea of AI as translator between forms. For the first time, GPT-5.5 has been trying to get me to tone these requests down so I don’t ruin my chances at the job.
PROMPT ENGINEERING
-
Best Claude Code Skills Found After 400 Hours of Testing
By
–
this guy literally spent 400 hours in Claude Code testing 100+ skills to find the 6 best ones 🤯
— Charly Wargnier (@DataChaz) 3 mai 2026
Nate put together an awesome guide and a 15-minute tutorial showing exactly how to make the most of them.
full video in 🧵↓ https://t.co/9Bnje8R1wo pic.twitter.com/M8695HPbWbthis guy literally spent 400 hours in Claude Code testing 100+ skills to find the 6 best ones Nate put together an awesome guide and a 15-minute tutorial showing exactly how to make the most of them. full video in ↓
-
Using Codex Agent to Hatch Pet from Reference Image
By
–
ask codex to /hatch pet based on this reference image
-
Better Prompting Styles Unlock Different AI Model Outputs
By
–
You don’t need a better AI model. You need a better prompting style. Structured → precision Analytical → accuracy Conversational → speed Planning → execution Same model. Different outputs. The model doesn’t change
Your prompting style does Follow -
Using Codex Chats for One-Off Tasks Without Project Pollution
By
–
chats in codex are underrated, you can do som many one off, sheets, docs, search etc tasks without worrying about it polluting your project more people should use chats in codex!!
-
Seeking Tips for Prompting Codex Goal Mode in Production
By
–
Going to try codex goal mode on a production-ish system, does anyone have notes on the right ways to prompt it?
-
Using GPT and Codex to Generate Docs Slides and Sheets
By
–
make sure you’re on the latest version of the app and with GPT 5.5 you should get good results across slides, sheets and docs what helps is to give some instructions to Codex about how the artefact should look like and then it vrooms past
-
ASCII Emoticons Can Trick LLMs Into Generating Harmful Code
By
–
Can smiley faces break AI? Researchers from Xi’an Jiaotong University, NTU, and UMass Amherst reveal a new LLM vulnerability: emoticon semantic confusion. ASCII emoticons like 🙂 can trick models into misinterpreting intent, leading to harmful code generation. Their study,
-
Gemini App Lacks Model Retry Feature Without Losing Attachments
By
–
Why doesn't the Gemini app let us retry prompts with a different model without losing our attachments? Doing the whole copy-paste and re-upload routine every time is a MASSIVE headache.
-
Stanford and UC Berkeley Present LLM-as-a-Verifier Framework
By
–
What if your LLM could verify its own reasoning with near-human precision? Stanford & UC Berkeley researchers present LLM-as-a-Verifier: a general-purpose framework that gives fine-grained feedback by breaking tasks into smaller criteria, scoring with higher granularity, and