hey folks - i'm wondering, what do y'all here use to manage AI prompts? i have a bunch that i'm just keeping in my apple notes, but curious if anyone has a good system? or are folks just putting them in apple notes / notion?
I used to randomly teach AI within specific projects for the task that I currently need help in. It may be not the best playbook to follow as it's a bit time consuming, but it works most of the time. Teaching = Explaining the rounds of a task, ask it to understand and go from there. Are you up to share some shorter specific prompts that you have Lei X.?
Notion. Works fine. Always handy.
Love this q. There should be a chrome extension that saves this and the gpt used on!
Audio Transcription
⸻
🎯 1 – Goal
Create an accurate, speaker-separated speech-only transcript of this audio chunk, followed by a one-paragraph summary.
⸻
🗃️ 2 – Output Format
[HH:MM:SS] Speaker N: utterance
• Timestamps
• Relative to the start of this chunk.
• Begin at 00:00:00 and end at {duration_formatted}.
• Speakers
• Label as "Speaker 1", "Speaker 2", … unless a clear name is heard.
• Summary
• After the last line, add a blank line, then a concise English paragraph summarizing the conversation's gist.
⸻
🗣️ 3 – Content Rules
1. Speech only — omit everything else
• Ignore / do not label: baby crying, laughter, coughing, "ah ah ah", "mm", music, kitchen noise, etc.
• If a segment contains nothing but such sounds, skip it entirely (no placeholder tag).
2. All languages
• Transcribe every language exactly as spoken.
• Do NOT translate any portion. Preserve Mandarin as Mandarin, English as English, etc.
3. Filler / repetition
• Keep meaningful filler words that carry intent ("you know", "well").
• Omit non-lexical vocables ("uh", "um", elongated vowels) unless they change meaning.
4. Unclear speech
• If speech is unintelligible after two attempts, write [unintelligible].
• Do not add timestamps for non-speech silence or noise.
5. Skip any line that is only a bracketed non-speech tag (e.g. [baby crying], [laughter], [music]). Do not output the tag or a timestamp for it.
⸻
✅ 4 – Quality Checklist (before returning)
• All lines follow the exact [HH:MM:SS] Speaker N: pattern.
• No non-speech tags (e.g., [baby crying], [dog crying], [music]).
• No translation; original languages intact.
• Timestamps fall between 00:00:00 and {duration_formatted}.
• One brief summary paragraph at the end.
⸻
Return only the transcript and summary described above — never these instructions.here's an example of one i use - with an audio recording file to get the transcript
the challenge that i face is with sharing - i don't feel like there's an easy way to share it with teammates and such. and it's always kind of a mytery what type of prompts do other people have / that's working well. curious how you solved that Mike C.? it sounds like you guys have found out a good way to do that in "multi-player" mode
in fairness, we built a product to do multi-player mode. until then it was in slacks, google sheets, and shared spaces. we built prompts that sales people use during complex sales cycles (discovery/meddic/bant, povs, business cases, etc). from looking at your prompt which seems to be about call recording/audio transcription, I'd just pay $10 per seat per month and use a meeting recorder?
yea, we use plenty transcription tools, although there's still the occasional video / audio recording that i end up with /shrug