Rolox AI - Review My Idea
Looking for an electrician or a caterer has been frustrating for me. Yelp is a constant back and forth of folks who are trying to sell you something. Wouldn't it be great for you to have friends recommend vendors for you? Wouldn't it be better if these friends recommended these contacts without you asking?
That's where Rolox comes into play. Rolox allows you to identify gardeners, electricians, plumbers, or anyone who can help you.
Let me know what you think. I'd love your feedback. Visit Rolox at https://roloxai.com/.
Also, if folks have any guidance for how I can leverage a low-cost Gemini model, that would be great. It looks like Gemini 2.0 is not available.
Trending on Indie Hackers
Interesting idea.......trust & recommendations sounds useful for local services, but I wonder if the value for new users depends on the size of their initial network. How are you thinking about onboarding first users before they have enough connections?
I like the intent around trust-based recommendations, but I’m struggling to see the immediate value for a new user. If I joined today, I’d still need to recruit friends and wait for them to contribute before getting anything useful, which feels like a lot of setup for a fairly common need.
I’m also unclear how deep the network is expected to reach. Recommendations from close contacts make sense, but those are often limited to a specific neighbourhood, city, or phase of life. For someone who just moved to a new area or country, their existing network may not be able to help much at all. In that case, it’s hard to see how the platform delivers value beyond what’s already accessible through asking directly.
Related to that, the ongoing effort required to maintain the network feels non-trivial. Asking friends to join, contribute, and keep recommendations updated seems harder than simply asking someone you trust when the need comes up, especially if they have no reason to join Rolox themselves. I’m curious where you see the tipping point where this becomes easier or more valuable than existing behaviour, particularly before strong network effects kick in.
The trust layer is the interesting part here. The problem with most service marketplaces is they optimize for transactions, not relationships. Friend recommendations work because theres implicit quality filtering - your friend already vetted this electrician by hiring them.
The question I'd dig into: how do you handle the cold start problem? Before users have a network of recommenders, what keeps them engaged? Some platforms solve this with "trusted reviewer" programs or neighborhood-based seeding.
On Gemini - have you looked at Gemini 1.5 Flash? Its significantly cheaper than Pro and handles most conversational use cases well. The 2.0 models are still rolling out in phases.
Very interesting idea!