TIL you can use open-source models in codex
AI
-
Should You Be Polite to AI Chatbots?
By
–
Should You Really Say Please And Thank You To AI? Millions of people now say please and thank you to #AI #chatbots, but does #politeness actually improve the answers or simply waste time and energy? As AI becomes more human-like, the bigger question may be how our
-
AI Agents as Orchestration Layers: Skills, MCP, Subagents, Hooks, Memory
By
–
AI agents are more than LLMs • Skills → expertise
• MCP → external connections
• Subagents → delegated tasks
• Hooks → automation
• Memory → persistent context Real AI systems are orchestration layers. Via Giuliano Liguori (
@ingliguori
) #AI #AIAgents #MCP -
Stop manual prompts, build systems that prompt themselves
By
–
STOP TYPING MANUAL PROMPTS INTO CLAUDE
— Charly Wargnier (@DataChaz) 17 juin 2026
In this 26-minute breakdown, an Anthropic engineer shares a massive mental model shift:
"You shouldn't prompt models, you should build systems that prompt themselves"
According to the breakdown, 95% of builders are missing the core… https://t.co/FEVhNDLHjS pic.twitter.com/gx4PZJhbPoSTOP TYPING MANUAL PROMPTS INTO CLAUDE In this 26-minute breakdown, an Anthropic engineer shares a massive mental model shift: "You shouldn't prompt models, you should build systems that prompt themselves" According to the breakdown, 95% of builders are missing the core
-
How messy internet data becomes a working model
By
–
We all know these models are trained on the internet. And the internet is full of spam, duplicates, toxic text, and personal data nobody should be memorizing. So how does any of that turn into a model that actually works? It doesn't go in raw. The data runs through a pipeline
-
Use AI as a support tool, not as a substitute for thought
By
–
La clave no es evitar la IA, sino usarla como una herramienta de apoyo y no como un sustituto del pensamiento pic.twitter.com/NAXaLvn7Eb
— Juan Merodio (@juanmerodio) 17 juin 2026The key is not to avoid AI, but to use it as a support tool and not as a substitute for thought.
-
One prompt turns AI into a technical co-founder for real apps
By
–
Almost every "build an app with AI" attempt ends the same way: a nice-looking mockup that breaks the second you click anything real. This one prompt fixes that. It turns ChatGPT, Claude, or Gemini into a technical co-founder that builds a real, working product with you across 5
-
Inverse scaling: weak model gains from harness patching runtime gaps
By
–
Exactly the inverse-scaling point. The weak model gained most because the harness patched gaps it couldn't fix on its own. Cheaper to evolve the runtime than to scale the model, and you keep the smaller footprint at inference.
-
Indian AI startups showcase global innovations at VivaTech 2026
By
–
What does AI innovation from India look like on the global stage?
— IndiaAI (@OfficialINDIAai) 17 juin 2026
At @VivaTech 2026, founders from Flaunt, InLustro and Daten & Wissen share how they are building AI solutions in trend intelligence, workforce readiness and enterprise computer vision—and taking them to the world… pic.twitter.com/xISma4lx5UWhat does AI innovation from India look like on the global stage? At @VivaTech 2026, founders from Flaunt, InLustro and Daten & Wissen share how they are building AI solutions in trend intelligence, workforce readiness and enterprise computer vision—and taking them to the world
-
Anthropic : désactivation de Fable 5 et Mythos 5 pour étrangers et action en justice
By
–
En une semaine de juin, deux événements se sont produits qui semblent sans lien mais racontent en réalité la même histoire : Le Département du Commerce a forcé Anthropic à désactiver Fable 5 et Mythos 5 pour tout ressortissant étranger, tandis que le DOJ est allé en justice au