Ask HN: Help me pick between two validated startup ideas | Hacker News Reader

Ask HN: Help me pick between two validated startup ideas

6 pointsiJohnPaul9 comments

For the last couple of months, I've been indie hacking on a couple of AI projects, just to see where long term value is. I'm currently exploring two ideas that might be long term potential but I don't know which one to commit to.

1. Hyperaide. https://hyperaide.com/. Middle layer for building AI applications. Having launched a few AI products over the past months, there are a few standout problems when it comes to prompt management and things. This aims to solve that with a focus on builders and startups as opposed to enterprise (Humanloop for the little guy lets say).

2. Chatlango. https://www.chatlango.com/. A platform to practice speaking languages by having conversations with AI personas and real users. Supplementary to other forms of learning. Consumer but has B2B side with selling to educational institutions.

Hyperaide launched last week. Few users but some are paying customers. Chatlango launched few weeks ago. ~1000 users. No paying customers.

What I like about Hyperaide: Its a dev tool. "selling shovels" might be worthwhile in the AI boom. Have experienced problems first hand. Has paying customers. What I like about Chatlango: It has a decent amount of users. Its consumer so its a bit "cool" to do. Also a personal problem. I am trilingual and language learning has been a big part of my life.

Been thinking about this for a few weeks but still on the fence.

Chatlango is going to be in a very crowded space. Kind of startup that requires running at a loss for a while. And since funding is not exactly what it used to be, it’s a gamble. Could generate some passive income though.

Hyperaide is going to also be in a crowded and fast moving space. But easier to sell/be profitable.

#2 feels like it would be easily destroyed if it got any traction by an incumbent offering something similar (Duolingo, Babbel etc).

#1 seems to have a better moat and more novel value. Plus people are paying for it already, which is a huge endorsement. I'd go for #1.

Do the first. Both are solutions I'm personally having trouble with, but the first is the one that I immediately want to throw my money at.

Pricing is a bit weird though. $20 is extremely cheap for unlimited, extremely expensive for a prototype.

Chatlango sounds like something you might not even have expertise in. It doesn't have the languages I want (Arabic, Indonesian). I'm not even confident that I'm fluent enough to hold a conversation with it or that it simulates the conversation accurately. Seems a bit like a one eyed leading the blind thing.

Would love to chat with you if possible. Just looking to learn more about what problems are most important for you regarding LLM dev tooling.

No 1, nobody needs another shitty ai chat app in a foreign language

As someone who built a React based AI foreign language chat partner using the davinci GPT3 model all the way back in 2022 purely for personal use, this space is already hyper competitive - it's not a defensible moat at all.

Which one do you see yourself having the motivation to work on for the next 10 years?

Is #2 based on #1's technology? If so, then you can use it as a test/demo app for #1.

#1 could be viable. #2 is a UI product and difficult to defend or get traction.

Reply on news.ycombinator.com