[Iter-X] 74/100days

Day 7️⃣4️⃣

First day of the holiday and I stayed at home, haha. Went out today and witnessed the sheer number of tourists — it was absolutely packed. Despite that, I still managed to release two new versions. High-efficiency output!

While I was out, I got inspired with a new idea: an insurance assistant. Right now, due to various reasons, the awareness and acceptance of insurance among many people in China is relatively low. There are several causes, but I believe one major reason is the resistance to salespeople and the headache of trying to understand complex insurance policies. The information gap also leads to major differences in perception. For example, today we encountered a question about whether a 70-year-old can still purchase critical illness insurance, and had to dig through specific policies to find the answer.

So I thought — what if we collect policy terms and product data from mainstream insurance companies, and use AI + RAG to build a digital insurance advisor? One that helps users quickly identify suitable products based on their needs, while explaining confusing terms in simple language. This would also help educate the market. From what I know, many insurance agents don’t have strong backgrounds — even though that’s improving, overall professionalism remains uneven.

This product could be offered for free, and monetized by partnering with insurance providers — earning commissions when users make purchases through the platform. I think this idea has strong potential for commercialization. I haven’t done formal market research yet (there are probably several competitors already), but I still believe the idea has value. If I choose to pursue it, I’ll likely start by building a version just for myself and close friends, then promote it organically in relevant communities. No monetization in the early phase — just focusing on solving real user pain points. I believe that could lead to a solid user base.

Have you ever tried an AI doctor that reads medical reports? I found it extremely helpful. In real life, many doctors don’t have the time or resources to give you thorough explanations. That’s why some patients become experts over time — they’re forced to. AI doctors can fill this gap well. While they might err on the side of caution in diagnosis, they’re great for objective education and personalized health advice. You can ask them step-by-step about a new illness, understand its mechanisms, and learn how your habits might contribute to it. That allows you to take targeted action. I believe the same model could work brilliantly in other vertical domains, including insurance.

Of course, building a good product is about removing what doesn’t work, but as a creator or entrepreneur, I think the mindset should be: first identify and amplify what works, then use your ability and resources to fix what doesn’t. Success might not be far from there. These are two different mindsets I often observe. I wonder — which type are you? There’s no absolute right or wrong — just a difference in what suits the path you’re on.

Current Progress Summary:

1. Prototype Design & UI/UX: 33%
2. Backend (Go): 60%
3. Client (Flutter): 58%
4. Data: 14%

If you meet the following criteria, feel free to reach out:

1. Can stay committed
2. Have dreams
3. Have passion