I met an AI lab that said their AI is now writing kernel code. So expect to see even that part of the OS eaten into.
AI
-
Top AI Stories: I/O 2026, Codex, Agents, Regulation
By
–
Top stories in AI today: – Exclusive insights from Sundar Pichai at I/O 2026
– OpenAI’s latest wave of Codex upgrades
– Generate an agent-native CLI from any website
– California moves to protect workers impacted by AI
– 4 new AI tools, community workflows, and more -
Ethics Must Be Central to AI Design, Not Afterthought
By
–
La ética en la ia no debería añadirse al final del proceso como un simple ajuste. Porque cuando se convierte en un parche, normalmente ya es demasiado tarde.
— Juan Merodio (@juanmerodio) 22 mai 2026
La innovación no consiste solo en crear herramientas más potentes, sino en diseñarlas con una visión humana pic.twitter.com/xVzzCQC0siEthics in AI shouldn't be added at the end of the process as a mere adjustment. Because when it becomes a patch, it's usually already too late. Innovation doesn't just consist of creating more powerful tools, but of designing them with a human vision
-
AI Math Tutors Will Soon Match Field Medalist Expertise
By
–
This might actually be the best time to get into math. Soon enough everyone will have a field medalist level math *tutor* at their disposal.
-
Quantifying Hyperparameter Transfer and Embedding Learning Rate in LLMs
By
–
Quantifying Hyperparameter Transfer and the Importance of Embedding Layer Learning Rate (first screenshot, Kalra and Barkeshli): https://
arxiv.org/abs/2605.21486 Optimal Embedding Learning Rate in LLMs: The Effect of Vocabulary Size (Hayou and Liu): https://
arxiv.org/abs/2506.15025 -
µP embedding LR rule correct under AdamW, explains most benefits
By
–
To clarify, this paper basically says: under AdamW, µP's embedding LR rule (constant) is essentially right and explains most of µP's benefit. Last year, Hayou et al. found that µP's embedding LR rule is wrong for realistic LLM vocab sizes. They found that the optimal embedding
-
Sonnet vs Opus: speed, cost, and when to switch
By
–
4. Sonnet 4.6 handles most tasks faster than Opus 4.7 at a fraction of the cost. Opus is the heavy lifter for complex reasoning, long code sessions, and deep analysis. But for everyday work? Sonnet gets it done in half the time. Default to Sonnet. Switch to Opus when the task
-
Claude features: Memory plus Search Past Chats
By
–
3. Memory + Search Past Chats work together. Memory lets Claude remember your preferences, projects, and context across conversations. Search Past Chats lets you ask "what was the pricing model I finalized last Wednesday?" and Claude pulls the actual conversation. Turn both on
-
Custom Claude style improved output quality
By
–
2. Custom Styles change output quality more than any prompt tweak. I have one called "skeptical senior operator" that pushes back on weak logic instead of agreeing with everything I say. Took 3 minutes to create. Single biggest quality jump I've gotten from Claude. Settings >
-
Persistent project context for AI chats
By
–
1. Projects hold your entire context permanently. Your brand guidelines. Your writing samples. Your past deliverables. Drop them in once. Every new chat inside that Project starts with full context. Stop re-explaining who you are and what you need in every conversation.