I think, seeing recordings of people using LLMs to accomplish non-trivial tasks would go a long way.
I’d love to watch, e.g. you Simon, using these tools. I assume there are so many little tricks you figured out over time that together make a big difference. Things that come to mind:
- how to quickly validate the output?
- what tooling to use for iterating back and forth with the LLM? (just a chat?)
- how to steer the LLM towards a certain kind of solutions?
- what is the right context to provide to the LLM? How do it technically?
Thanks for all the detailed responses. Very much appreciated food for thought.
I have a specific question about the calculation of the fair value. My understanding is that there are two steps:
1. Project the future earning (that is the difficult part)
2. Discount the earnings and get the NPV - the fair value of the company
Currently the interest rates are so low that the NPV will be quite high and not that far from the current valuations. Indeed, this seems to be the rationalization by many for the sky high valuations of today. But this seems wrong to me. What would you advice? Pick some other interest rate, maybe an aspirational rate of return for ones investing? Or don't try to compute the NPV and think in terms of multiples like P/E and P/S?
Part of the problem is that you don't have time to think deeply about things, because there is to much to think about. The solution is simple: drop almost everything except what is most important to you and allocate plenty of time to for it. (This solution is simple but not easily implemented.)
A good read on this topic is "The Shallows: What the Internet Is Doing to Our Brains" by Nicholas Carr
I’d love to watch, e.g. you Simon, using these tools. I assume there are so many little tricks you figured out over time that together make a big difference. Things that come to mind:
- how to quickly validate the output?
- what tooling to use for iterating back and forth with the LLM? (just a chat?)
- how to steer the LLM towards a certain kind of solutions?
- what is the right context to provide to the LLM? How do it technically?