It's important to recognize that there are no optimal "right" answers. If someone is telling you there is a best way of doing this, always treat their advice with suspicion. (Yes, this includes what I'm going to write below.) The reason is that because AI advances so quickly, there isn't enough time for the industry to stabilize on best practices and spread them widely before the underlying system changes and it's no longer applicable.
Because of this, I find it very important that you build your own understanding of LLMs, and create your own best practices from these first principles. Then, when you read about the next best thing, you can decide if it makes sense or not. Is it hype? Is it real? Does it logically align with what I understand about LLMs?
That being said, here's two things I find universally applicable:
1. You have to get very good at asking questions. Not only that, you also have to ask questions about the questions you should be asking. You also have to ask it to ask YOU questions.
2. Spec driven development is, in my opinion, a good place to start. Writing down what you're going to do and how you're going to do it has always been a good practice in any industry.
Spend time thinking through your daily work routine and create a list of toil items. Try to describe the toil and steps you usually take when tackling it and write them out in a markdown file. Tell your model of choice where to find the file and start engaging in a two way conversation with it.
So I am taking a different approach than instead of just focusing on software. I have Gemini app, Claude app, Grok app and then chatgpt still using web interface. And periodically I ask them questions about what I can do to improve the wellbeing of humans. I think trying to use them for work is okay, but it's not the best way to use these tools, we can use them to broaden our thinking and find challenges we never considered before.
My approach is to try delegating as many daily tasks to AI as possible. For example, I've now completely replaced my read-it-later tool with my agent. I just send him links, and he periodically summarizes the content I've sent him.
It's important to recognize that there are no optimal "right" answers. If someone is telling you there is a best way of doing this, always treat their advice with suspicion. (Yes, this includes what I'm going to write below.) The reason is that because AI advances so quickly, there isn't enough time for the industry to stabilize on best practices and spread them widely before the underlying system changes and it's no longer applicable.
Because of this, I find it very important that you build your own understanding of LLMs, and create your own best practices from these first principles. Then, when you read about the next best thing, you can decide if it makes sense or not. Is it hype? Is it real? Does it logically align with what I understand about LLMs?
That being said, here's two things I find universally applicable:
1. You have to get very good at asking questions. Not only that, you also have to ask questions about the questions you should be asking. You also have to ask it to ask YOU questions.
2. Spec driven development is, in my opinion, a good place to start. Writing down what you're going to do and how you're going to do it has always been a good practice in any industry.
Spend time thinking through your daily work routine and create a list of toil items. Try to describe the toil and steps you usually take when tackling it and write them out in a markdown file. Tell your model of choice where to find the file and start engaging in a two way conversation with it.
Go from there.
thanks!
So I am taking a different approach than instead of just focusing on software. I have Gemini app, Claude app, Grok app and then chatgpt still using web interface. And periodically I ask them questions about what I can do to improve the wellbeing of humans. I think trying to use them for work is okay, but it's not the best way to use these tools, we can use them to broaden our thinking and find challenges we never considered before.
My approach is to try delegating as many daily tasks to AI as possible. For example, I've now completely replaced my read-it-later tool with my agent. I just send him links, and he periodically summarizes the content I've sent him.
[dead]
[dead]
[dead]