What processes do we have? What are the specific tasks?
Posted: Sun Feb 16, 2025 8:18 am
Decomposing tasks
To create some process optimization, the most important thing to do is to decompose the task. From experience, I can say that this is the biggest problem at this stage.
Process optimization
This is where most of the time is spent. But it is china telegram data important to emphasize that when setting the context and providing data, you sometimes need to consider the concept of a context window. Neural networks do not have infinite memory, so you cannot load 1000 PDF files into it. Instead, five A4 pages are enough.
It is important not to overload the system with unnecessary information, as it analyzes text, images, and graphics very qualitatively. After receiving the first result, always provide feedback: indicate what you like, what you don't, what needs to be improved, and start the process again.
The main goal is to create an effective prompt that you can rely on. For example, if a new employee joins your company in the marketing department and the task is to generate a weekly newsletter, you already have a ready-made prompt that completes this task in two minutes. The new employee does not need to be explained where to go and what to do.
Competitor analysis can also be optimized. There are many examples, but the advantage of neural networks is that they are applicable to all areas and departments, especially in the pharmaceutical industry.
To create some process optimization, the most important thing to do is to decompose the task. From experience, I can say that this is the biggest problem at this stage.
Process optimization
This is where most of the time is spent. But it is china telegram data important to emphasize that when setting the context and providing data, you sometimes need to consider the concept of a context window. Neural networks do not have infinite memory, so you cannot load 1000 PDF files into it. Instead, five A4 pages are enough.
It is important not to overload the system with unnecessary information, as it analyzes text, images, and graphics very qualitatively. After receiving the first result, always provide feedback: indicate what you like, what you don't, what needs to be improved, and start the process again.
The main goal is to create an effective prompt that you can rely on. For example, if a new employee joins your company in the marketing department and the task is to generate a weekly newsletter, you already have a ready-made prompt that completes this task in two minutes. The new employee does not need to be explained where to go and what to do.
Competitor analysis can also be optimized. There are many examples, but the advantage of neural networks is that they are applicable to all areas and departments, especially in the pharmaceutical industry.