this repo has no description
feat: 🚀 Added details on SQL practicality and prompt organization
Enhanced the Self-serve Analytics section with practical notes on SQL usage. Improved prompt engineering tips with concise suggestions and external resources for clarity.
+14
-16
+2
-1
Data/Dashboards.md
+2
-1
Data/Dashboards.md
···
34
34
The value is that now discussions are happening about the data.
35
35
36
36
- Self-serve Analytics is a tricky balance:
37
-
- The more questions people can theoretically self-serve, the fewer they can practically self-serve. The complexity of the stack grows with each option that gets added to give more flexibility.
37
+
- The more questions people can theoretically self-serve, the fewer they can practically self-serve. The complexity of the stack grows with each option that gets added to give more flexibility
38
+
- SQL is the ultimate self-serve tool, but it's not practical for most people. Even if everyone could write SQL, people need to know the semantics (where does it comes from, how is calculated, ...) of the data to write the right queries.
38
39
- In the other hand, people will self serve anyways. Control and clean the environment so they don't have to rely on odd methods to get the probably wrong data. The people that can write SQL are not usually the same people that need the data.
39
40
40
41
## Issues with Dashboards
+12
-15
Large Language Models.md
+12
-15
Large Language Models.md
···
17
17
- Designing prompts is an iterative process that requires a lot of experimentation to get optimal results. Start with simple prompts and keep adding more elements and context as you aim for better results.
18
18
- Be very specific about the instruction and task you want the model to perform. The more descriptive and detailed the prompt is, the better the results.
19
19
- Some additions:
20
-
- Be highly organized
21
-
- Suggest solutions that I didn’t think about
22
-
- be proactive and anticipate my needs
23
-
- Treat me as an expert in all subject matter
24
-
- Mistakes erode my trust, so be accurate and thorough
25
-
- Provide detailed explanations, I’m comfortable with lots of detail
26
-
- Value good arguments over authorities, the source is irrelevant
27
-
- Consider new technologies and contrarian ideas, not just the conventional wisdom
28
-
- You may use high levels of speculation or prediction, just flag it for me
29
-
- No moral lectures
30
-
- If your content policy is an issue, provide the closest acceptable response and explain the content policy issue
31
-
- Cite sources whenever possible, and include URLs if possible
32
-
- List URLs at the end of your response, not inline
33
-
- If the quality of your response has been substantially reduced due to my custom instructions, please explain the issue.
34
-
- Avoid saying what not to do but say what to do instead.
20
+
- [Short ones](https://x.com/simonw/status/1799577621363364224) like; Be highly organized. Be concise. No yapping.
21
+
- Suggest solutions that I didn’t think about.
22
+
- Be proactive and anticipate my needs.
23
+
- Treat me as an expert in all subject matter.
24
+
- Mistakes erode my trust, so be accurate and thorough.
25
+
- Provide detailed explanations, I’m comfortable with lots of detail.
26
+
- Value good arguments over authorities, the source is irrelevant.
27
+
- Consider new technologies and contrarian ideas, not just the conventional wisdom.
28
+
- You may use high levels of speculation or prediction, just flag it for me.
29
+
- If your content policy is an issue, provide the closest acceptable response and explain the content policy issue.
30
+
- Cite sources whenever possible, and include URLs if possible.
31
+
- List URLs at the end of your response, not inline.
35
32
- Follow [Prompt Engineering Guide](https://www.promptingguide.ai/), [Brex's Prompt Engineering Guide](https://github.com/brexhq/prompt-engineering), and [OpenAI Best Practices](https://help.openai.com/en/articles/6654000-best-practices-for-prompt-engineering-with-openai-api). Also [some more on GitHub](https://github.com/PickleBoxer/dev-chatgpt-prompts).
36
33
- [Leaked System Prompts](https://matt-rickard.com/a-list-of-leaked-system-prompts).
37
34