Author Preface: Why This Book

Over the past few years, large language models have given me a simple but powerful insight:

An intelligent system’s output depends not only on how capable it is, but also on the context it is placed in.

In the beginning, many of us treated LLMs like chatbots. Ask one question, get one answer. Sometimes the model felt brilliant. Sometimes it felt vague, confident, or wrong. Naturally, we focused on the model itself: model size, training data, reasoning ability, coding ability, mathematical ability.

Then we began to notice another variable.

Context.

Give a model a vague request, and it usually gives a vague answer. Give it a clear goal, background, constraints, examples, tools, memory, and evaluation criteria, and it may move from “chatting” to “working.”

The same model can feel like two different systems under different contexts.

This has a deep implication for life.

If we think of a human life as a complex intelligent system, then life output is not shaped only by talent, effort, or willpower. It is shaped by the context a person lives inside: what they read, what they hear, who they spend time with, what organization they enter, what space they live in, what tasks train them, what feedback they receive, and what story they believe about themselves.

We often imagine life change as a dramatic decision.

But from the evolution of LLMs, we can see a quieter truth: stable change often comes from rewriting context.

To manage life well is not to burn willpower every day. It is to carefully manage what feeds you, what shapes you, what trains you, what drains you, and what calibrates you.

In one sentence:

To reboot your life system, start by managing your life context.