Almost never comment on vids, i js wanted to show my gratitude for this tho. it feels like a year of learning about chatgpt summarized in 20 minutes in a way that any1 can understand. we need more tutorials like this.
🎯 Key Takeaways for quick navigation: 00:00 📝 *Introduction to ChatGPT Prompt Engineering Guide* - Overview of the official prompt engineering guide by OpenAI for better results with ChatGPT. - Simplification of strategies for both developers and everyday users. - Highlighting the importance of providing clear instructions to AI models. 01:09 🎯 *Strategy: Write Clear Instructions* - Emphasis on the need to detail prompts with context for better results. - Example: Using specific details in questions rather than generic queries. - Advising against relying on one-sentence prompts due to AI limitations. 02:31 🎭 *Strategy: Adopt a Persona* - Recommending the use of personas to enhance AI responses. - Example: Instructing the model to include jokes in responses. - Application of personas for different scenarios and backgrounds. 03:53 🖋️ *Strategy: Use Delimiters for Clear Input* - Explanation of delimiters (quotation marks, XML tags) for refining output. - Application for complex prompts and maintaining a specific format. - Enabling more precise instructions as prompts become more intricate. 05:16 📝 *Strategy: Specify Steps for Task Completion* - Importance of specifying steps to guide AI in completing tasks. - Example: Breaking down tasks into clear steps for effective results. - Addressing the challenge of copying entire pages without step guidance. 06:26 🔄 *Strategy: Provide Examples to Model Responses* - Utilizing examples to instruct AI on the desired response format. - Encouraging users to guide AI by presenting sample answers. - Example: Teaching AI to respond in a philosophical style. 07:35 📏 *Strategy: Specify Desired Output Length* - Advising users to specify the length of desired AI responses. - Recommendation: Opting for paragraphs or bullet points for accuracy. - Noting variations in AI response accuracy based on length instructions. 08:18 📚 *Strategy: Provide Reference Text* - Guiding users to instruct AI to answer using specific reference text. - Example: Using health articles to derive answers in a given context. - Application for incorporating external sources or custom knowledge bases. 09:28 🗂️ *Strategy: Split Complex Tasks into Subtasks* - Recognizing the need to break complex tasks into manageable subtasks. - Example: Using intent classification to identify relevant instructions. - Highlighting the importance of addressing complex tasks systematically. 11:19 🔄 *Strategy: Summarize Previous Dialogues* - Addressing the challenge of limited AI memory in dialogue applications. - Recommending summarization of previous dialogues for context. - Example: Asking AI to provide a brief summary of the prior conversation. 13:55 🧩 *Strategy: Piecewise Summarization for Long Documents* - Managing the context length limitation by summarizing documents in parts. - Example: Summarizing specific chapters of a book to overcome constraints. - Acknowledging the need to adapt to AI limitations in handling longer texts. 16:01 ⏳ *Strategy: Give Models Time to Think* - Encouraging users to allow AI time to work out solutions independently. - Example: Teachers asking AI to check student solutions systematically. - Incorporating inner monologue to control the thought process in responses. 18:07 🛠️ *Strategy: Use External Tools for Knowledge Retrieval* - Geared towards developers, introducing the use of external tools for knowledge bases. - Mention of the API cookbook and embedding knowledge bases into API. - Brief overview of actions, functions, and API calls for advanced users. 19:01 🔄 *Strategy: Test Changes Systematically* - Emphasizing the importance of systematic testing when making prompt changes. - Example: Testing and comparing changes to prompts for better results. - Highlighting the need for tracking and analyzing prompt adjustments. 19:55 🧪 *Strategy: Test Changes Systematically (Continued)* - Continuing the discussion on systematic testing for prompt changes. - Stressing the significance of monitoring prompt modifications for effectiveness. - Recognizing the iterative nature of prompt engineering for optimal results. 20:54 🧪 *AB Testing & Systematic Changes* - Systematically test prompt changes for better results. - Test multiple changes and ensure repeatability. - Save successful prompts for future use. 21:51 📊 *Summary & Free Resources* - Summary of the six prompt engineering strategies. - Offer of a downloadable PDF with simplified prompts. - Invitation to explore Skill Leap AI platform with free resources. 22:19 🚀 *Skill Leap AI Platform* - Introduction to Skill Leap AI platform. - Over 15 AI courses available, covering various tools. - New courses released monthly with a free 3-day trial offer. Made with HARPA AI
The strategies and techniques discussed in the RUclips video transcript for improving prompts in ChatGPT can be concisely listed as follows: 1. **Clear Instructions**: - Include details in questions for more relevant answers. - Adopt a persona in prompts, like asking for a response with a joke or in the style of a business consultant. - Use delimiters to clearly indicate distinct parts of the input, such as triple quotation marks or XML-like tags. 2. **Reference Text**: - Instruct the model to answer using reference text, like using a specific article to answer a question. - Answer with citations from a reference text. 3. **Complex Tasks into Subtasks**: - Use intent classification to identify the most relevant instructions. - Summarize or filter previous dialogue in long conversation scenarios. - Summarize long documents piece-wise and construct a full summary. 4. **Model's Time to Think**: - Instruct the model to work out its own solution before concluding. - Use inner monologue to allow the model to process thoughtfully. - Ask the model if it missed anything in the previous pass for self-review. 5. **Use of External Tools** (primarily for developers): - Employ external tools and APIs for better knowledge retrieval. - Systematically test changes in prompts for optimal results. 6. **Test Changes Systematically**: - Change one aspect of the prompt at a time and compare results. - Save effective prompts for repeated use. Each strategy encompasses various tactics aimed at refining the interaction with ChatGPT to achieve more accurate and relevant responses.
🎯 Key Takeaways for quick navigation: 00:00 📝 *Strategia 1: Scrierea de instrucțiuni clare* - Detalii în întrebare pentru răspunsuri relevante. - Solicitați modelul să adopte o persoană. - Utilizați delimitatoare pentru claritate. 01:09 📚 *Strategia 2: Furnizarea de texte de referință* - Instruiți modelul cu citări din text de referință. 10:23 🧩 *Strategia 3: Descompunerea sarcinilor complexe* - Utilizați clasificarea intențiilor. - Rezumați sau filtrați dialogul extins. - Sumarizați documente pe porțiuni. 15:49 🤔 *Strategia 4: Acordarea de timp modelelor* - Modelul să lucreze la soluție înainte de concluzie. - Utilizați monolog intern pentru confidențialitate. - Cereți modelului să verifice rezultatele. 19:01 🛠️ *Strategia 5: Utilizarea uneltelor externe* - Folosiți unelte externe pentru recuperare mai bună. - Implementați apeluri de API externe. - Explorează acțiuni avansate pentru interacțiuni complexe. 20:26 🧪 *Strategia 6: Testarea schimbărilor sistematic* - Salvați prompturile eficiente. - Testați modificările sistematic. - Utilizați teste A/B pentru comparații. 21:51 📊 *Analiza și Salvarea Rezultatelor Testelor* - Efectuați teste multiple pentru cele mai bune modificări. - Salvarea rezultatelor în prompt books. - Asigurați repetabilitate și gestionare ușoară. 22:19 📘 *Accesarea Resurselor Suplimentare și Platformei Skill Leap AI* - Descărcați un PDF cu prompturi și strategii. - Skill Leap AI oferă cursuri complete despre AI. - Acces la resurse și cursuri avansate. Made with HARPA AI
🎯 Key Takeaways for quick navigation: 00:14 🤖 *Simplifying OpenAI's official prompt engineering guide for non-developers to get better ChatGPT results * 00:55 💡 *Strategy 1: Give clear instructions with details and context to get more relevant ChatGPT answers* 02:19 🎭 *Tactic: Ask ChatGPT to adopt a persona or style to shape its responses * 03:53 📋 *Tactic: Use delimiters like triple quotes to separate distinct input parts * 04:35 📎 *Tactic: Specify step-by-step instructions for completing complex tasks * 06:26 📑 *Tactic: Provide example responses to guide ChatGPT's output style* 07:35 ✂️ *Tactic: Set desired output length for more accurate summaries * 08:18 📚 *Strategy 2: Provide reference texts for ChatGPT to cite * 10:23 🧩 *Strategy 3: Split complex tasks into simpler chatbot-style subtasks * 15:49 ⏱️ *Strategy 4: Give ChatGPT models time to evaluate solutions * 18:07 🔎 *Tactic: Ask ChatGPT to check its work and provide additional details* 20:26 📊 *Strategy 6: Test prompt changes systematically to optimize results* Made with HARPA AI
1:00 I think they did release it then or it's certainly been around for a long time. Perhaps they have just updated recently but I read through the guides many months ago.
You are amazing thank you so much!!! Is there a pdf guide of prompts and setup (like those that you use here) on your skill leap? I downloaded the resources and am not finding one like you mention. Thank you so much!! 🙏🙏🙏🙏
Thank you. On skill leap membership, there are a ton of downloads for members. We have prompt books and PDF downloads for all our courses there. You can email me if you don’t find a specific one saj@skillleap.ai
Thank you for the great lesson. I have a question about adopting a persona, can i use it also when building my own GPT and give that instruction at the configuration page? or i need to add that at the beginning of each new chat?
Totally off the subject. I'm sorry. I am trying to buy an item on OfferUp. I have a question. How do I access the entire thread of my messages with the seller? I can only see the last half. Thanks. Again, sorry for hijacking this post.
![Chlöe - Shake (Feat. Jeremih) [Official Visualizer]](http://i.ytimg.com/vi/aDs_1ufpfv8/mqdefault.jpg)
Master ChatGPT, Midjourney, and top 50 AI tools with Our New AI Education Platform.
Start a free trial Today: bit.ly/skillleap
Almost never comment on vids, i js wanted to show my gratitude for this tho. it feels like a year of learning about chatgpt summarized in 20 minutes in a way that any1 can understand. we need more tutorials like this.
Thank you! really appreciate this
Welcome 2 a New Era of Prompt Engineering
ruclips.net/user/shortsok-UekYjSs8?feature=share
🎯 Key Takeaways for quick navigation:
00:00 📝 *Introduction to ChatGPT Prompt Engineering Guide*
- Overview of the official prompt engineering guide by OpenAI for better results with ChatGPT.
- Simplification of strategies for both developers and everyday users.
- Highlighting the importance of providing clear instructions to AI models.
01:09 🎯 *Strategy: Write Clear Instructions*
- Emphasis on the need to detail prompts with context for better results.
- Example: Using specific details in questions rather than generic queries.
- Advising against relying on one-sentence prompts due to AI limitations.
02:31 🎭 *Strategy: Adopt a Persona*
- Recommending the use of personas to enhance AI responses.
- Example: Instructing the model to include jokes in responses.
- Application of personas for different scenarios and backgrounds.
03:53 🖋️ *Strategy: Use Delimiters for Clear Input*
- Explanation of delimiters (quotation marks, XML tags) for refining output.
- Application for complex prompts and maintaining a specific format.
- Enabling more precise instructions as prompts become more intricate.
05:16 📝 *Strategy: Specify Steps for Task Completion*
- Importance of specifying steps to guide AI in completing tasks.
- Example: Breaking down tasks into clear steps for effective results.
- Addressing the challenge of copying entire pages without step guidance.
06:26 🔄 *Strategy: Provide Examples to Model Responses*
- Utilizing examples to instruct AI on the desired response format.
- Encouraging users to guide AI by presenting sample answers.
- Example: Teaching AI to respond in a philosophical style.
07:35 📏 *Strategy: Specify Desired Output Length*
- Advising users to specify the length of desired AI responses.
- Recommendation: Opting for paragraphs or bullet points for accuracy.
- Noting variations in AI response accuracy based on length instructions.
08:18 📚 *Strategy: Provide Reference Text*
- Guiding users to instruct AI to answer using specific reference text.
- Example: Using health articles to derive answers in a given context.
- Application for incorporating external sources or custom knowledge bases.
09:28 🗂️ *Strategy: Split Complex Tasks into Subtasks*
- Recognizing the need to break complex tasks into manageable subtasks.
- Example: Using intent classification to identify relevant instructions.
- Highlighting the importance of addressing complex tasks systematically.
11:19 🔄 *Strategy: Summarize Previous Dialogues*
- Addressing the challenge of limited AI memory in dialogue applications.
- Recommending summarization of previous dialogues for context.
- Example: Asking AI to provide a brief summary of the prior conversation.
13:55 🧩 *Strategy: Piecewise Summarization for Long Documents*
- Managing the context length limitation by summarizing documents in parts.
- Example: Summarizing specific chapters of a book to overcome constraints.
- Acknowledging the need to adapt to AI limitations in handling longer texts.
16:01 ⏳ *Strategy: Give Models Time to Think*
- Encouraging users to allow AI time to work out solutions independently.
- Example: Teachers asking AI to check student solutions systematically.
- Incorporating inner monologue to control the thought process in responses.
18:07 🛠️ *Strategy: Use External Tools for Knowledge Retrieval*
- Geared towards developers, introducing the use of external tools for knowledge bases.
- Mention of the API cookbook and embedding knowledge bases into API.
- Brief overview of actions, functions, and API calls for advanced users.
19:01 🔄 *Strategy: Test Changes Systematically*
- Emphasizing the importance of systematic testing when making prompt changes.
- Example: Testing and comparing changes to prompts for better results.
- Highlighting the need for tracking and analyzing prompt adjustments.
19:55 🧪 *Strategy: Test Changes Systematically (Continued)*
- Continuing the discussion on systematic testing for prompt changes.
- Stressing the significance of monitoring prompt modifications for effectiveness.
- Recognizing the iterative nature of prompt engineering for optimal results.
20:54 🧪 *AB Testing & Systematic Changes*
- Systematically test prompt changes for better results.
- Test multiple changes and ensure repeatability.
- Save successful prompts for future use.
21:51 📊 *Summary & Free Resources*
- Summary of the six prompt engineering strategies.
- Offer of a downloadable PDF with simplified prompts.
- Invitation to explore Skill Leap AI platform with free resources.
22:19 🚀 *Skill Leap AI Platform*
- Introduction to Skill Leap AI platform.
- Over 15 AI courses available, covering various tools.
- New courses released monthly with a free 3-day trial offer.
Made with HARPA AI
The strategies and techniques discussed in the RUclips video transcript for improving prompts in ChatGPT can be concisely listed as follows:
1. **Clear Instructions**:
- Include details in questions for more relevant answers.
- Adopt a persona in prompts, like asking for a response with a joke or in the style of a business consultant.
- Use delimiters to clearly indicate distinct parts of the input, such as triple quotation marks or XML-like tags.
2. **Reference Text**:
- Instruct the model to answer using reference text, like using a specific article to answer a question.
- Answer with citations from a reference text.
3. **Complex Tasks into Subtasks**:
- Use intent classification to identify the most relevant instructions.
- Summarize or filter previous dialogue in long conversation scenarios.
- Summarize long documents piece-wise and construct a full summary.
4. **Model's Time to Think**:
- Instruct the model to work out its own solution before concluding.
- Use inner monologue to allow the model to process thoughtfully.
- Ask the model if it missed anything in the previous pass for self-review.
5. **Use of External Tools** (primarily for developers):
- Employ external tools and APIs for better knowledge retrieval.
- Systematically test changes in prompts for optimal results.
6. **Test Changes Systematically**:
- Change one aspect of the prompt at a time and compare results.
- Save effective prompts for repeated use.
Each strategy encompasses various tactics aimed at refining the interaction with ChatGPT to achieve more accurate and relevant responses.
I love that we can specify paragraph counts instead of word counts to meet limit requirements. Great stuff!
thanks bro this video helps me a lot ❤🙏, love from Bharat(India)
Can't get enough DOPE ChatGPT insights into proper Prompt Engineering techniques. Thanks always Skill Leap AI for the latest info!
Welcome 2 a New Era of Prompt Enigneering
ruclips.net/user/shortsok-UekYjSs8?feature=share
A great vid.. No need for a paid course.. A million thanks!
Thank you, Thank you. I truly appreciate you putting this together !!!
Glad it was helpful!
Love from India ❤
i love absolutely everything about this video, thank you for inspiring my studying of this powerful and important topic with these great nuances.
A huge thanks for explaining and sharing👌This is just awesome😊
Glad you liked it!
🎯 Key Takeaways for quick navigation:
00:00 📝 *Strategia 1: Scrierea de instrucțiuni clare*
- Detalii în întrebare pentru răspunsuri relevante.
- Solicitați modelul să adopte o persoană.
- Utilizați delimitatoare pentru claritate.
01:09 📚 *Strategia 2: Furnizarea de texte de referință*
- Instruiți modelul cu citări din text de referință.
10:23 🧩 *Strategia 3: Descompunerea sarcinilor complexe*
- Utilizați clasificarea intențiilor.
- Rezumați sau filtrați dialogul extins.
- Sumarizați documente pe porțiuni.
15:49 🤔 *Strategia 4: Acordarea de timp modelelor*
- Modelul să lucreze la soluție înainte de concluzie.
- Utilizați monolog intern pentru confidențialitate.
- Cereți modelului să verifice rezultatele.
19:01 🛠️ *Strategia 5: Utilizarea uneltelor externe*
- Folosiți unelte externe pentru recuperare mai bună.
- Implementați apeluri de API externe.
- Explorează acțiuni avansate pentru interacțiuni complexe.
20:26 🧪 *Strategia 6: Testarea schimbărilor sistematic*
- Salvați prompturile eficiente.
- Testați modificările sistematic.
- Utilizați teste A/B pentru comparații.
21:51 📊 *Analiza și Salvarea Rezultatelor Testelor*
- Efectuați teste multiple pentru cele mai bune modificări.
- Salvarea rezultatelor în prompt books.
- Asigurați repetabilitate și gestionare ușoară.
22:19 📘 *Accesarea Resurselor Suplimentare și Platformei Skill Leap AI*
- Descărcați un PDF cu prompturi și strategii.
- Skill Leap AI oferă cursuri complete despre AI.
- Acces la resurse și cursuri avansate.
Made with HARPA AI
Thanks for these great videos, I'm going to buy that course.
Amazing as always! thank you!!
🎯 Key Takeaways for quick navigation:
00:14 🤖 *Simplifying OpenAI's official prompt engineering guide for non-developers to get better ChatGPT results *
00:55 💡 *Strategy 1: Give clear instructions with details and context to get more relevant ChatGPT answers*
02:19 🎭 *Tactic: Ask ChatGPT to adopt a persona or style to shape its responses *
03:53 📋 *Tactic: Use delimiters like triple quotes to separate distinct input parts *
04:35 📎 *Tactic: Specify step-by-step instructions for completing complex tasks *
06:26 📑 *Tactic: Provide example responses to guide ChatGPT's output style*
07:35 ✂️ *Tactic: Set desired output length for more accurate summaries *
08:18 📚 *Strategy 2: Provide reference texts for ChatGPT to cite *
10:23 🧩 *Strategy 3: Split complex tasks into simpler chatbot-style subtasks *
15:49 ⏱️ *Strategy 4: Give ChatGPT models time to evaluate solutions *
18:07 🔎 *Tactic: Ask ChatGPT to check its work and provide additional details*
20:26 📊 *Strategy 6: Test prompt changes systematically to optimize results*
Made with HARPA AI
Which tool do you use to make this list, do I use any GPT? What was the prompt?
Thank you for the answer
tks excellent !
1:00 I think they did release it then or it's certainly been around for a long time. Perhaps they have just updated recently but I read through the guides many months ago.
You are amazing thank you so much!!! Is there a pdf guide of prompts and setup (like those that you use here) on your skill leap? I downloaded the resources and am not finding one like you mention. Thank you so much!! 🙏🙏🙏🙏
Thank you. On skill leap membership, there are a ton of downloads for members. We have prompt books and PDF downloads for all our courses there. You can email me if you don’t find a specific one saj@skillleap.ai
Sir, to say the least, You're The MOST!! Just sayin'.
Thank you 👍🏻
Alright!
thanks
nice:)
Thank you for the great lesson. I have a question about adopting a persona, can i use it also when building my own GPT and give that instruction at the configuration page? or i need to add that at the beginning of each new chat?
No with your own GPT, you give it enough instructions so it doesn’t need these kind of prompts.
Thank you@@SkillLeapAI
Totally off the subject. I'm sorry. I am trying to buy an item on OfferUp. I have a question. How do I access the entire thread of my messages with the seller? I can only see the last half. Thanks. Again, sorry for hijacking this post.
If no one answers or offers a response, you're not hijacking the post.
13:23 Summarize the previous answer, After a New Prompt is given.
(They don't have long memory)
Great video. I would love to know which application you used to create the video itself. The sounds and presentation graphics are very clean. Thanks!
Thanks. It’s all done inside Adobe premiere pro.
Lotsa work there. Nice job.@@SkillLeapAI
Welcome 2 a New Era of Prompt Engineering
ruclips.net/user/shortsok-UekYjSs8?feature=share
links doesnt work, can you please renew or send me
Can you send me an email and I'll attached it to the PDF to the reply. support@skillleap.ai
👍👍✨🙏🇮🇳🙏