Took a couple of days off the internet, touched the grass. The ability to think without a model in the loop may become more important than we realize.
Every time I do this, it becomes harder to justify. With this acceleration, two days offline feel like two weeks.
When you stop processing five different terminal outputs at once, your brain gets to do something it forgot how to do: think slowly, for longer stretches, without reaching for a tool. I really believe we'll see some meaningful research emerging in this area soon.
Outside of the AI bubble, more people are starting to sense that something big is coming, but very few are talking about it seriously. As more people say, it's giving early Covid energy.
And thanks to @philipkiely for the signed Inference Engineering copy. I’m learning a lot from it.
Does anyone else get exhausted constantly re-explaining your specific business logic (schema, metric definitions) to LLM every time you start a new SQL session? 🤯
❓How do YOU currently handle this "Context Amnesia" with LLMs?
#DataEngineering#SQL#BuildInPublic#DataAnalytics
Context Engineering Skills 10x'd my project creation.
I rebuilt my digital brain system with Claude Code using the Context Plugin, so it is now a Personal OS.
It provides a complete folder-based architecture for managing:
- Personal Brand - Voice, positioning, values
- Content Creation - Ideas, drafts, publishing pipeline
- Knowledge Base - Bookmarks, research, learning
- Network - Contacts, relationships, introductions
- Operations - Goals, tasks, meetings, metrics
Because we're using Context Engineering principles in the Skills & Agent [.]md files, the system can be easily managed by small agents.
I'm adding the entire repo to GitHub tonight 🤫
Some people who succeed wildly in school don’t achieve their apparent potential in the business world. Some others who do okay (or worse) in school manage to build an extremely successful life. Why is that?
What we learn in school & must unlearn in business & in life:
(1/10)