GPT-6 Astra模型分步迭代优化技巧
This works really well with GPT-6 Astra:
This works really well with GPT-6 Astra:
这与 GPT-6 Astra 配合效果极佳：
Give it a tweet of an impressive Astra demo.
给它发一条关于令人印象深刻的 Astra 演示的推文。
Ask Astra (medium) to replicate it to the best of its ability and giving it whatever extra instructions and adaptations you want. Set a /goal like provide proof of the results so it has something to compare to.
要求 Astra（medium）尽其所能复制该演示，并根据需要添加任何额外的指令和调整。设置一个 /goal，例如提供结果证明，以便它有所对比。
After the first run, switch to Astra (max) and give it instructions to polish. And you can keep doing this iteratively to keep improving results.
第一次运行后，切换到 Astra (max) 并给出优化指令。你可以反复迭代进行，以持续提升结果。
So there is one component to build and one to optimize/tune.
因此，这里有一个组件用于构建，另一个用于优化/调优。
I think it works well because it breaks the problem down and allows the models to focus efforts as opposed to trying to use lots of tokens for many things at once (usually lower quality results).
我认为这种方法行之有效，因为它将问题分解开来，使模型能够集中精力，而不是试图一次性使用大量 token 处理许多事情（通常会导致质量较低的结果）。
This can essentially be done in one go using subagents and /goal.
这本质上可以通过子代理和 /goal 一次性完成。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力