Also, the signup flow is pretty painful. I have to upload a document, then pay, then go to my email and click on a link, that sends me another link to my email to sign in, then I complete the sign in, and it shows me the order information. And then no where does it let me chat with the robot about my doc...
That UX flow seems strange because the screen flow should just be: 1) upload, 2) pay, 3) chat. It shouldn't have asked you to check your email for anything.
Please feel free to email me at alfarez at gmail dot com if you're still seeing this issue.
As for handling large files, I'm using a popular method of breaking down the doc into small chunks and then just sending the relevant chunks to the AI.
How to pick the "relevant" chunks? Use embeddings and search for the embeddings that are "semantically closest" to the question being asked.
I think it's supposed to use the microphone to record your answers possibly? But it's not doing that for me either. I just get 15 seconds, and then "you failed."
It forced me to login, and I hate logging into stuff, so I didn't try it. But I like the idea. Another wrapper for GPT? Did you do any tweaking to the model? I think I've seen that developers can add training data to the model to give it more context or something"? Did you do any of that?
It was completely open originally, but I had to put it behind a login in order to limit how many times a user hit the OpenAI API.
I have creativity controls which adjust the "temperature", as well as revision controls which give you options to revise sections of the generated cover letter
Some feedback:
Firstly, this is loading full res photos on the homepage, which making the site extremely slow to load. You should generate some small thumbnails so that loading times aren't so terrible when looking at these gallery pages.
I would love more of a satirical approach. I would design it to be a very professional looking website, maybe knock off christie's or the louvre's design or something. Maybe have bio's about the authors (obviously keep it safe, because they're kids). And jack up the prices to be 100x what they currently are. Do really professional sounding descriptions for each piece (maybe you can use chatgpt to do this). "Art STUFF, is a still life Simeon. It's use of bright color pastel contradicts the subject matter of the piece and causes the viewer to consider the actual tools of the artist versus what the stereotypical tools that Simeon finds being pushed onto him." Or whatever...
Maybe professionally frame the pieces (or photoshop them to look professionally framed).
This seems like a fun side project. I'm also in the kid's market. I'd might be interested in partnering in some way if you're open to it.
Thanks for the tip! Everyone's been complaining about google recently about how the results are all just SEO garbage now. I could easily see ChatGPT become the goto for asking programming questions.
I pay for a budget app. https://www.monarchmoney.com/ But somehow I got in on their beta, and felt like they did a good job and wanted to continue to support them. But they've been around for quite a few years now, so I think they are pretty successful..
I was hoping smarter people than me would have responded by now; I apologize, because it kind of looks like you're only going to get me.
One strategy would be to write a blog post about how google and twitter won't approve your ads because they are too gay. Post this to Hacker News and a relevant subreddit, and hope that picks up some traction and that someone at google has the power to make things right.
After that, it's probably just keep doing what everyone else does. If you have a budget, you could try reddit ads, snapchat ads, instagram ads, etc. Although, you might have similar problems at those places as well. And you might consider hiring an ad agency that specializes in dating sites (if there are any). I work at a company that sells firearm accessories and we also had a very hard time with paid advertising.
You could go to youtubers directly, and sponsor some videos. Maybe you could find some youtubers that will partner with you and be willing to do more than just the typical sponsored segment in a video.
You could do SEO activites, like content writing for various keywords that you might be able to rank for. I'm still trying to figure this out myself, but people seem to have great success with it.
You might just try to look at what other dating sites (especially dating sites targeting LGBT) are doing/did in the early days for inspiration. I hear stories about how tinder went from campus to campus manually signing up women at all the sororities.
If you don't have a budget, then you're really really fighting an uphill battle, as a dating site is the kind of thing that needs a CONSTANT influx of users, and you can't really rely the occasional viral post if it's followed by 4 months of quiet. My guess is that you're going to need to do a bunch of things that don't scale well, while at the same time trying to raise money so that you can do more of the things that do scale.
Putting up a landing page is useless, unless you are going to actively drive traffic to it. And those leads are going to convert much much better if you can deliver some value on that page.
I put up a landing page 4 years before I had my product ready to go (a baby book I was making), and over four years, I gathered about 200 email addresses. Only a fraction of those convert, and they were probably worth $500 at the end of the day. Once I had the book finished and could show people how it worked, I dwarfed that by 10x in one day.
You can build a landing page. But I wouldn't let it distract you too much from the actual project, unless you are great at marketing, in which case, play to your strengths.
