I just use whatever is available, yday that was Grok 4.1 Fast Non Reasoning, now I try Grok 4.3 via grok-latest with reasoning set to none
PROMPT ENGINEERING
-
Anthropic’s Playbook on Claude’s AI Prompt Engineering
By
–
🚨 Anthropic quietly dropped a FREE 33-page playbook on Claude’s ultimate cheat code:
— Charly Wargnier (@DataChaz) 8 mai 2026
the 'Skills' folder ✨
Take 30 mins to set it up and stop repeating yourself forever.
Pros build systems, not prompts.
link below ↓ pic.twitter.com/9z2x2stON7Anthropic quietly dropped a FREE 33-page playbook on Claude’s ultimate cheat code: the 'Skills' folder Take 30 mins to set it up and stop repeating yourself forever. Pros build systems, not prompts. link below ↓
-
Post-training: Task-Specific but Less Realistic
By
–
Post-training is a double-edged sword: it makes the model more suited to a task, but less connected to reality.
-
Thousands of Vibe-Coded Apps Expose Corporate and Personal Data Online
By
–
Thousands of Vibe-Coded Apps Expose Corporate and Personal Data on the Open Web | WIRED https://
share.google/jakCcHeK2SW64e
K2i
… #viben #vibecoding #coding #coder #AI #anthropic #artificialintelligence @AlbertoEMachado @Eli_Krumova @postoff25 @Khulood_Almani @anand_narang @NutritiousMind -
GPT 5.5 excels in refactoring with e2e tests
By
–
/goal + GPT 5.5 is amazing. I can now plan really extensive refactors with e2e tests and it just works.
-
Abacus AI’s ChatLLM: One Subscription for Multiple Top AI Models
By
–
Stop paying for 5 different AI subscriptions.
— Abacus.AI (@abacusai) 7 mai 2026
ChatLLM by Abacus AI puts GPT-5.5, Claude 4.7, Gemini 3.1, DeepSeek-V4, Kimi-2.6 in one place.
One prompt → routed to the best model → output.
Pay once. Use all of them. pic.twitter.com/I2yqpvEDZ8Stop paying for 5 different AI subscriptions. ChatLLM by Abacus AI puts GPT-5.5, Claude 4.7, Gemini 3.1, DeepSeek-V4, Kimi-2.6 in one place. One prompt → routed to the best model → output. Pay once. Use all of them.
-
DeepAgents Series: Tools, MCP, and Skills
By
–
Kwasi is back for the deepagents series @DataScienceDojo
. Tools, MCP, and skills, what they are, how they work together, and how to build agents that can actually interact with the world. May 13 | 11am PT https://
datasciencedojo.com/event/ai-agent
-tools-mcp-skills/
… -
Open Source Real-Time World Models Enable Interactive Control
By
–
Open source real-time world models legit feel like puppeteering reality itself.
— Bilawal Sidhu (@bilawalsidhu) 7 mai 2026
Now @reactorworld is putting it in everyone’s hands at sub 50 ms latency
— type a prompt, change the scene, and control what unfolds as it generates. pic.twitter.com/99PObbdqqbOpen source real-time world models legit feel like puppeteering reality itself. Now @reactorworld is putting it in everyone’s hands at sub 50 ms latency — type a prompt, change the scene, and control what unfolds as it generates.
-
Copy prompts for instant coding with Codex
By
–
best part about the blogpost is that you can copy a prompt and directly get shipping with codex and gpt-realtime-2 you can just codex things!!
-
Better prompts create better AI outputs with structured frameworks
By
–
Better prompts = better AI outputs • Structured → precision
• Analytical → research
• Conversational → ideas
• Planning → execution Great prompts are frameworks, not guesses. Via Giuliano Liguori (
@ingliguori
) #AI #Prompts #GenAI