Would you use a deep research app for crypto trading as analysis tool?

[anywaifu]

So I am working in a deep research app, and I can see LOTS of use cases. I am not going to use stripe or any other payment gateways for now. Crypto is going to be the only payment method and I think it might hurt my profits, but it is so easier to implement that I will just use that for a while.

My idea was to provide support for people using the app through discord, answering questions and helping to create research plans. It will have a 20$ plan and a 50$ plan. these plans give user credits which can be spent for researches. It is similar to google deep research and openai deep research. The difference is that you can create your own plans with different capabilities and audit them. It is not as complex as n8n because it is totally focused in deep research, so you will find only components related to it like, query generator, query templates, web search, news search, twitter search, instagram search, blockchain search.

What I was thinking: user wants to buy an asset that just launched, but he wants to first know who is the team, what are people talking about it in twitter? instagram, what is the blockchain activity etc, then user creates a plan that can receive a query for a single asset and it will start the research that goes to all of these places and gather documents (text chunks from sources like web pages tweets etc). In the end, it is connected to a "text synthesizer" that will tell a story in the configured format of what was collected.. One could also use the "agent component" which will iterate over your prompt and search whatever you want automatically, but it is a bit more expensive due to reasoning and less accurate as llm might just hallucinate lol.

Does it make sense at all? would you pay $20 for it? I mean, I am trying to bring price as low as possible per research. it will depend on the complexity of the research. To be fair, it could even be upgraded in the future to have, instead of a "text synthesizer" also a "trading signal synthesizer" which runs periodically a set of rules, search specific pages, blockchain activity, indicators... It could also use its memory with graphrag capabilities and decide if triggers a signal or not. I think that deep research is a field yet to be explored. there are so many niches! since I will only be accepting crypto in the beginning, I thought it would make sense to ask here first if it really make sense or not..

The application is already online, but I am not advertising it yet

[rat03gopoh]

Does it make sense at all? would you pay $20 for it?

This makes sense, but is it worth $20-$50? Users have to do some testing, right?
The NeuroBro terminal allows free users to submit any crypto-related prompts, including in-depth analysis of TA or FA versions, with a limit of three prompts per day. Based on my testing, it has an accuracy rate of up to 70%. If you're creative enough, you can get 3 answers in 1 prompt.
I actually wanted to prompting it based on what you wrote here and see if NeuroBro could provide a comprehensive response. Unfortunately, my prompt limit has been reached, so I'll try again tomorrow.

[anywaifu]

Does it make sense at all? would you pay $20 for it?

This makes sense, but is it worth $20-$50? Users have to do some testing, right?
The NeuroBro terminal allows free users to submit any crypto-related prompts, including in-depth analysis of TA or FA versions, with a limit of three prompts per day. Based on my testing, it has an accuracy rate of up to 70%. If you're creative enough, you can get 3 answers in 1 prompt.
I actually wanted to prompting it based on what you wrote here and see if NeuroBro could provide a comprehensive response. Unfortunately, my prompt limit has been reached, so I'll try again tomorrow.

I am not sure how it works, how its agent acts. I asked it to generate a report about Pendle Boros for example, and it just created a summary. From what I can see, it is an agent specifically made for helping finding trading opportunities right? I think it is not really comparable to my app because it is mainly for deep research. for example, I made https://github.com/anywaifu/research-outputs/blob/main/pendle-boros.md using my deep research app. I am not trying to create a coin for it or anything similar. I am just accepting crypto as payment. The $20 is for an user to buy credits in order to make his researches. Gemini Ultra offers 200 researches by day while Open Ai $200 plan offers 120 per month. This is a brutal difference, but OpenAI is far superior. That's where I am in: I will try to make a cheap deep research while making reports similar to the OpenAI style or even superior

[SFR10]

What I was thinking: user wants to buy an asset that just launched, but he wants to first know who is the team, what are people talking about it in twitter? instagram, what is the blockchain activity etc, then user creates a plan that can receive a query for a single asset and it will start the research that goes to all of these places and gather documents (text chunks from sources like web pages tweets etc). In the end, it is connected to a "text synthesizer" that will tell a story in the configured format of what was collected..[...] Does it make sense at all? would you pay $20 for it? I mean, I am trying to bring price as low as possible per research.

If I were to enter this field as someone who had limited knowledge, I would've tried using a service like this ^{[until I become comfortable with my own skills]}, but probably a cheaper one ^{[assuming the cost you mentioned earlier was for a single asset]}.
^{- IMO, those who have super ultrawide monitors can easily do it by themselves after a few weeks (it's a bit harder to keep track of stuff on normal monitors).}

[anywaifu]

What I was thinking: user wants to buy an asset that just launched, but he wants to first know who is the team, what are people talking about it in twitter? instagram, what is the blockchain activity etc, then user creates a plan that can receive a query for a single asset and it will start the research that goes to all of these places and gather documents (text chunks from sources like web pages tweets etc). In the end, it is connected to a "text synthesizer" that will tell a story in the configured format of what was collected..[...] Does it make sense at all? would you pay $20 for it? I mean, I am trying to bring price as low as possible per research.

If I were to enter this field as someone who had limited knowledge, I would've tried using a service like this ^{[until I become comfortable with my own skills]}, but probably a cheaper one ^{[assuming the cost you mentioned earlier was for a single asset]}.
^{- IMO, those who have super ultrawide monitors can easily do it by themselves after a few weeks (it's a bit harder to keep track of stuff on normal monitors).}

The idea is to keep the price as low as possible, when doing a deep research, there are many AI services and data providers involved for doing things like google search, website data extraction, twitter extraction etc.. You still use your knowledge but to do research on the topics you want. The idea was to compete against most big deep research providers like open ai and good while allowing users to customize the research process. 20usd could be able to do 10 or 200 researches. all depends on the research complexity user is running.. that's the proposal of the app

[The Cryptovator]

So I am working in a deep research app, and I can see LOTS of use cases. I am not going to use stripe or any other payment gateways for now. Crypto is going to be the only payment method and I think it might hurt my profits, but it is so easier to implement that I will just use that for a while.

Good that you are interested in something that could help crypto users, but it should be something unique from existing tools. So crypto users would be interested to use it. Since you intend to add a crypto payments gateway, it will definitely be helpful for those who are interested in your service or tools. I am still not sure how users are going to evaluate your tools to make a payment.

Does it make sense at all? would you pay $20 for it? I mean, I am trying to bring price as low as possible per research. it will depend on the complexity of the research. To be fair, it could even be upgraded in the future to have, instead of a "text synthesizer" also a "trading signal synthesizer" which runs periodically a set of rules, search specific pages, blockchain activity, indicators... It could also use its memory with graphrag capabilities and decide if triggers a signal or not. I think that deep research is a field yet to be explored. there are so many niches! since I will only be accepting crypto in the beginning, I thought it would make sense to ask here first if it really make sense or not..

The application is already online, but I am not advertising it yet

It depends on your service; either it's worth paying for or not. Users should have a trial period, like three days; they could use it for free, and then if they were interested, they would proceed with payment. It's better to keep some service for free always, for premium users need to pay. Once everything is ready, you can create your own announcement thread with all the details of your tools. So the user might check all the terms.

[anywaifu]

So I am working in a deep research app, and I can see LOTS of use cases. I am not going to use stripe or any other payment gateways for now. Crypto is going to be the only payment method and I think it might hurt my profits, but it is so easier to implement that I will just use that for a while.

Good that you are interested in something that could help crypto users, but it should be something unique from existing tools. So crypto users would be interested to use it. Since you intend to add a crypto payments gateway, it will definitely be helpful for those who are interested in your service or tools. I am still not sure how users are going to evaluate your tools to make a payment.

Does it make sense at all? would you pay $20 for it? I mean, I am trying to bring price as low as possible per research. it will depend on the complexity of the research. To be fair, it could even be upgraded in the future to have, instead of a "text synthesizer" also a "trading signal synthesizer" which runs periodically a set of rules, search specific pages, blockchain activity, indicators... It could also use its memory with graphrag capabilities and decide if triggers a signal or not. I think that deep research is a field yet to be explored. there are so many niches! since I will only be accepting crypto in the beginning, I thought it would make sense to ask here first if it really make sense or not..

The application is already online, but I am not advertising it yet

It depends on your service; either it's worth paying for or not. Users should have a trial period, like three days; they could use it for free, and then if they were interested, they would proceed with payment. It's better to keep some service for free always, for premium users need to pay. Once everything is ready, you can create your own announcement thread with all the details of your tools. So the user might check all the terms.

There is a free tier in place already.. It is very limited though. It is a service that require too many third party services to operate which makes things a bit expensive, but yes, people will be able to use it for free

[rat03gopoh]

I am not sure how it works, how its agent acts. I asked it to generate a report about Pendle Boros for example, and it just created a summary.

Yeah, I don't know what the premium version is like either (where the connected wallet should hold around $4,000 worth of tokens at current rates), but I'm sure you can do even crazier things since there are no limits; just adjust the prompts.

So, let's focus on your platform. What is the end product for users? Not a summary either?
Too much story telling is certainly not good, as it's the same as your tool not doing the work that streamlines the process. The results of deep research from your sophisticated tools and how the final analysis should be the product you offer here, so users don't have to do that anymore.

Despite some naysayers, AI has been credited with super-capabilities, even becoming a personal financial advisor. The fewer words an AI produces that encompass comprehensive research and analysis, the more desirable it is.

[retaur]

Does it make sense at all? would you pay $20 for it?

This makes sense, but is it worth $20-$50? Users have to do some testing, right?
The NeuroBro terminal allows free users to submit any crypto-related prompts, including in-depth analysis of TA or FA versions, with a limit of three prompts per day. Based on my testing, it has an accuracy rate of up to 70%. If you're creative enough, you can get 3 answers in 1 prompt.

Does it have access to the internet and its own datasets or could it work as well as a local instance of a big text and image to text AI model? I'm wondering if what it runs gives a similar output to what open source models can deliver (if you've got an example you can DM me, I'll give it a whirl on two sizeable models I've got downloaded).

So I am working in a deep research app, and I can see LOTS of use cases. I am not going to use stripe or any other payment gateways for now. Crypto is going to be the only payment method and I think it might hurt my profits, but it is so easier to implement that I will just use that for a while.

If it's a website then fine, if it's an app (like what you're expecting someone could install on an iPhone or Android device, you might be struggling as the main app stores of both likely won't allow an app that accepts only crypto purchases).

I think you'd need to try to take a more dynamic pricing model than just having a $20/query idea. If you're able to take a current model and train it up to be extremely precise and extremely crypto specific you'd probably get something pretty good but I'm not sure whether AI models compete with both auditors and hft algorithms (high frequency traders/market makers).

[rat03gopoh]

Does it have access to the internet and its own datasets

Yes it does and I don't know about it
I'm just a free user and only have very limited access (send 3 simple prompts). If you just want to see what their output is like in the free version and how it compares to other open-source tools, you're better off trying it yourself (G-search keyword: "Neuro terminal")