Throwback to this to-do app I prototyped where physical notes would create inputs in mixed reality. Your tasks and reminders would sit mapped to your wrist and oriented based on time. Might be worth revisiting with the VisionPro!

Think different. #applevisionpro

Apple vs AI hardware startups

New hobby unlocked - watching my flight descent on Apple Watch

even the sora mistakes are mesmerizing

😂

When they ask me to QA the app

Worried about sending sensitive documents to ChatGPT? I built an app that automatically identifies and redacts sensitive personal information using a custom trained AI model that runs locally on your device. Using Cursor . Cursor helped me… - Write a complete Electon

1/ Here's a secret: you hear this a lot because it's true. Most incubator/accelerators and pre-seed VCs expect to see traction even when they say otherwise. If not revenue then at least user growth. Why the doublespeak?

Investing in Vertical AI Means Learning the Job

Excited to share LoveFrom and OpenAI have come together to form io, a new company creating the next generation of AI products and interfaces. I’ve rarely felt so inspired by a project and team! openai.com/sam-and-jony/

Is the employee Jony Ive

Whoever wins AI personalization wins the consumer. Nobody actually wants to work with multiple agents. Why everyone isn't scrambling to try and get more data and context from users right now to do better personalization is beyond me.

+1 for "context engineering" over "prompt engineering". People associate prompts with short task descriptions you'd give an LLM in your day-to-day use. When in every industrial-strength LLM app, context engineering is the delicate art and science of filling the context window

We've raised Structured Outputs limits to support larger schemas: Object properties: from 100 → 5,000 Characters in string: 15,000 → 120,000 Enum values: 500 → 1,000 Total characters across string members in enums with >250 values: 7,500 → 15,000

notes on context engineering with Walden of Cognition

We're building a visual canvas for designing prompts. This interaction makes it easy to test prompting concepts simply by dragging and dropping new sections into your existing prompts.

Returning to the US after a week in Europe