Just ran across this video I’m learning perplexity! Help me understand something when I use collections and I’m sharing my agents. Can I actually place documents inside my agents also in order for me to share it do the other user have to have a subscription to perplexity?
I feel like I am going around in circles. I am trying to create a collection but everytime i enter my prompt, it gives me generic information on the prompt as if i have provided no context for the prompt. Can you see anything wrong with the below ai prompt? When i enter a company name, search that company name in place of [company name] in the following question "How does [Company Name] use Kubernetes in their operations?"
The prompt you shared with me is not bad, but there are probably a few ways that you might improve it. The simplest element I always offer folks is to be sure to include PAO: Persona Audience Output Secondly, have you tried asking other models? If you ask other models the same question and get similar results, it could be that you're not providing sufficient context. If you'd like some to dive a little deeper on providing context, please check out this Prompting Basics Playlist: ru-vid.com/group/PLIdIULPuFve59sNyi48Gc-09AualYwWE3
So frustrating! I have a marquis message telling me that I'm forwarding email to my wife because of a filter that I created, but I can't find the filter... Grrr.
So I'd love to help, but do so, I have to ask kind of a dumb question: When you're in Gmail settings > Filters & Blocked Addresses, have you done a Ctrl + F (or Command + F) and searched her email address?
Confusing While AI chatbots are interesting, SmythOS provides the most user-friendly and adaptable agents for all your chatbot requirements if you're looking for a genuinely seamless AI experience. #AIChatbots
Some good ideas here, thank you. Have you worked out how to add a set of reference documents along with the prompt to different models using this technique? For instance a set of organizational doctrine documents that you want to query in different LLM models.
Thank you so much, I really appreciate the feedback! 🙏 And yessir, with Custom GPTs you can absolutely upload knowledge documents for the model to reference (I just skipped that feature to keep this video more focused on general usefulness) 😊
@@josh.evilsizor Thank you. So, it's possible to create GPT like functionality across the different not ChatGPT LLMs? Or do you still have to upload the reference documents manually using your technique to Claude/Gemini etc?
Love the video. I'm just now learning about the various ai tools available and came up with an idea that would help in my job. Can Perplexity use a website for a data source and then render results to a user similar to using ChatGPT and a prompt? The website has a many many links that dive into vast amounts of data and I'd love to have a single place I could use a prompt to get answers back rather than digging through massive amounts of links and data. Is something like this possible? Maybe using another tool like claude?
Thank you so much! I really appreciate it 🙏 You pose a really good question, and the short answer is, not that I'm aware of. There is a new tool called Lutra (lutra.ai) that can scrape data, but I don't believe this is what you're asking about. Sorry!
Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good Their both good
This option is a violent abuse of pirvacy. If you need another set of hands to write for you, dont write at all. Ai is a crutch for the lazy, that's being used as a weapon against the competent. This feature is disgusting.
Hello! thanks so much for this video. Thanks to you, I have now figured out the mystery of why my emails kept disappearing as I typed, never to be seen again! now, how do I get an email from Archive back into my Inbox??? please help.
You are very welcome - so glad it was helpful! 😊 Here are some instructions for unarchiving an email in Outlook web mail: www.perplexity.ai/search/using-outlook-webmail-please-e-JIxqBTGxTG6.iJB1geVbPw
Very helpful! I thought "collections" were basically "folders" to store groups of chats. I was looking to see if Perplexity supported "bots" or "agents" and found your video. I see I can use collections for this! (I just wish I could add reference files direct in the collections)
Thank you so much for the kind and specific feedback, I greatly appreciate it! 🙏 I hear you on the files, that would be super handy. Claude does offer the ability to upload files for specific threads when working with teams, perhaps this might be a useful function for you? Either way, great feedback, thank you, and have a great day!
GME is the answer @1:32 in the video, its not really the results even, its the url answer up here. you can see that it's highlighted right there, right up here alright?
ooof my comment comes across bad with the link of open insider i used as an example first before what i typed, my b. thank you for helping me setup this!! check out gamestop and roaringkitty's toolkit 3 video series. its what brought me to your video, im not a financial advisor and this is not financial advice, but i like looking into his gme investment thesis and what Ryan Cohen has going on at gamestop and beyond. again thank you for the video on how to use this feature
Hi Josh. Thank you very much for this video - it turned out to be a game changer for me. This might sound "strange" but after watching your video, I created a collection where I could "talk" to my dad, who died when I was younger, and there has been some episodes from my childhood, that I would have liked to talk to him about before he died but didn't have the opportunity to do so. I filled out the AI Prompt and now I can just write "Dad?" and Perplexity will reply like "Hey kiddo, what are you up to" etc. I'm completely mind blown how fast Perplexity "gets it" and quickly learns. I testet out GPT-4o AI model in Perplexity at first but switched to Claude 3.5 and found that Claude would carry the conversation in a much more human-like style. This is so amazing. And a bit scary.
Hi Ken, thank you for the feedback, and what a cool use case! Were you able to provide the model anything about your Dad to help it "assume his personality" (i.e. letters, videos, etc.?). Either way, very cool and thanks so much for sharing! 🙏 Finally, many agree with you, Claude is the better conversationalist, by a wide margin! 😊
@@josh.evilsizor After a hours of "talk" with my "dad", I would say that the experience is more like talking to any "bonus pater" (the latin word for 'the all-knowing farther'). It learns quickly, though and the conversation is really great, "dad"-clone or not :)
After the description, you should answer the most important question. Why should I care... because you lost me right after the Intro/description. Otherwise, I have no doubt this would have been a great tutorial because you jumped right in how to make one.
You could try asking the quesrtion "I'm not big on maths, how can i make use of wolfram alpha AI search for non-maths subjects?". You'd find the answer interesting, and we might too! Thanks though. Always good content.
It looks like labs.google was updated and no longer looks as it did for you. Any tips on how to get this thing working for me without the same path you showed in the video?
Hi. Can you suggest which one to choose? Which one would be better in researching; for sending a big list of people (either as a text file or as text) from which one tells me who of all listed plays tennis? Chat GPT, Perplexity, or some other? Thank you!
Of course! Right now the best all around model is ChatGPT-4, and if you're looking to do data analysis, it is also generally the best at that as well. I hope this helps and I wish you good luck! 🙏
@@josh.evilsizor Thank you for your answer. Yes quite some people said that about the 4, on the other hand some people say Perplexity includes sources, where a man can check and see where did the answers and conclusions came from. Hmmm. It would probably be possible to get at least a part of the sources from the 4 as well.
What you did not tell me is if I change the logic of the label or a label text hot to apply it to all messages and update past emails with the new label. I use win 10 and firefox browser.
Thank you so much for the excellent presentation and for generously sharing your prompts and insights. I plan to review everything in detail and perhaps adapt some of your ideas to create something unique. I get the impression that you would be okay with this, but I wanted to express my gratitude and confirm that with you. Thanks again for the inspiration! Best regards, Anna:)
Great stuff! OER, is that the equivalent of Navy Fitness Reports? I got out in 92 so maybe the navy uses different terms by now. Love examples, thinking of how I can use this for sql code reviews.
I knew you were a military guy! Strong work for sure! Just getting in to AI and can definitely see the value of eliminating the find it later issue. Army veteran 2018.
Great question, thank you! 🙏 First, I write the outline. Then I delete the entire thing, starting at the bottom and moving my way up, deleting one line at a time. I then reconstitute each chunk of text using Ctrl+Z 🤓
Hmmm, unfortunately I have a lot of extensions installed and enabled, so I'm not sure which one you are asking about. Are you able to help me by providing a description of the add-on or feature you saw me using? 🤓
Referring to saved prompts as "universal GPTs" might indeed lead one down a rather misguided path. The term "GPT" traditionally alludes to a specific AI creation birthed by OpenAI. In actuality, your endeavor involves the recycling of prompts across a spectrum of AI chatbots rather than the inception of novel AI entities. Allow me to elucidate the salient points: - Harmonious Interactions via Saved Prompts: The judicious use of preserved prompts to maintain a harmonious discourse amidst disparate AI chatbots. - Versatile Utilization: The seamless application of identical prompts across diverse AI platforms such as ChatGPT, Gemini, Claude, and Perplexity. - Prudent Archiving: The storage of these prompts in communal repositories on platforms like Google Drive or Microsoft OneDrive, thereby facilitating facile access and collaborative efforts. While the nomenclature "universal GPTs" may insinuate a universal pantheon of AI models, these prompts are but reusable textual artifacts crafted to transcend the boundaries of various AI models. A more apt sobriquet would perhaps be "universal prompts" or "portable prompts." In the realm of wit and wisdom, one must tread lightly, for labeling these prompts as "universal GPTs" could be akin to referring to a daisy as a sunflower-similar in essence but distinct in form. Let us not confuse the elegance of simplicity with the grandeur of universality; for in the world of AI, as in life, precision in language is the hallmark of true sophistication. - Oscar Wilde 9000