Built for this world so you can create the next. Now available in Jade, with Dune on the Pro Flex Keyboard.
All-day battery. Built-in AI. Performance that moves with you.
Learn More: https://t.co/zDKAKd42XM
Meet the new Surface RTX Spark Dev Box 🔥. Every detail of this experience was intentional, from purposeful configuration to preinstalled tools. We can't wait to see what you build with it. https://t.co/cqymZv4cpW
Introducing @Surface RTX Spark Dev Box, a compact developer PC engineered with NVIDIA RTX Spark silicon and built on the Windows developer platform, designed for local-first AI development. #MSBuild
Introducing - Surface RTX Spark Dev Box.
All the power developers need, right out of the box.
Sign up to learn more: https://t.co/DF0DvEvhOz
Surface RTX Spark Dev Box. Built for more.
Introducing Surface Laptop Ultra.
Built for world makers. Designed for what's next.
The most powerful Surface laptop ever. Coming Fall 2026.
Sign up to learn more: https://t.co/k8aEX2pTAy
How to become expert at thing:
1 iteratively take on concrete projects and accomplish them depth wise, learning “on demand” (ie don’t learn bottom up breadth wise)
2 teach/summarize everything you learn in your own words
3 only compare yourself to younger you, never to others
A @GitHubCopilot CLI skill that I use everyday:
Runs multi-model ensemble code reviews on your branch.
Why @GitHubCopilot CLI over standalone @claudeai or @OpenAICodexCli ?
Copilot gives you access to all the models in one place. Instead of picking one and hoping it catches everything, you can pit Claude, GPT, and Gemini against the same diff simultaneously and cross-validate their findings. You can also run the reviews in parallel using agents.
How it works: → Pick 2+ models (Claude Opus, Sonnet, GPT Codex, Gemini Pro, etc.) → Each reviews your diff in parallel across 5 dimensions: breaking changes, security, bugs, design, code quality → Findings are deduplicated & consolidated into one severity-ranked report
If 3/3 models flag the same bug, you know it's real. If only one model catches something, it's still surfaced — no blind spots.
https://t.co/4h9xgnw8VM
Interact to clean your timeline:
Transformers
LLMs
Prompt engineering
CoT
Constitutional AI
Ray Kurzweil predictions
TPU pods
Attention is all you need
noam shazeer
Scaling laws
Colab Pro
ooms
GPT wrappers
Model distillation
AGI timelines
p doom
NVDA VRT TSMC
open weights
unhobblings
infinite context
stargate
As far as I can tell, one of the biggest changes across organizations over the past few years is simply the rise of distraction. The default often appears to be a kind of continuous partial attention. I'm not sure whether it's good or bad. Maybe people were suboptimally stuck before, and perhaps there are gains to being able to allocate attention more flexibly? Or maybe it's bad because meetings become even less efficient and our ability to focus (already under assault!) withers further. Either way, it's a striking new normal.