The Story Behind SureLM
SureLM started with a conversation with an insurance agency owner who was struggling to train and retain his sales force. We initially thought this was a problem faced by individual agencies. But when we dug deeper, we found something much bigger.
A huge population of people who could potentially become effective insurance sellers—auto drivers, domestic workers, local professionals and others with strong community relationships—remain outside the insurance workforce because they lack the knowledge and training needed to do the job.
We then looked beyond India and found that the underlying challenge is global: insurance companies everywhere struggle to build, train and support large enough distribution networks.
That led us to a simple idea: what if the expertise of an experienced insurance salesperson could be replicated through AI?
We started building SureLM around this idea. The platform uses AI to understand customer needs, retrieve relevant product information, guide recommendations and support the salesperson through the process. As we built and tested it, we found that this approach could actually replicate parts of the knowledge and workflow that experienced sellers use.
Today, we are speaking with grassroots salespeople and the response has been encouraging. They understand the problem immediately and are excited about having this kind of support.
Our biggest challenge now is not whether the technology can work. It is figuring out who pays for it and proving its value in real-world deployments. We need pilots with institutions to demonstrate the impact, and connecting the technology into existing bank and insurance distribution networks requires their participation and permission.
That is the stage we are at now: a working idea, early validation from the people who would use it, and the next challenge of proving that it can work at scale.
Built With
- keycloak
- next.js
- node.js
- nomic-embed-text
- ollama
- postgresql
- prisma
- qdrant
- qwen-2.5-7b
- react