Bootstrapping a Multi-Agent RAG EdTech platform: 2 engines done, looking for seed funding.
Hi everyone! We are a two-person engineering team at Quantum Code G&S. Over the last 3 months, we've built a high-load B2B/B2C EdTech platform for the driving school market entirely from scratch.
What we have successfully built and packed:
1. An Advanced Web Platform powered by a hybrid RAG (Retrieval-Augmented Generation) system.
2. A High-Availability asynchronous Telegram Bot that acts as a smart pocket instructor.
Both core engines are fully finished, localized for legal database parsing, and ready in our private repos.
The current bottleneck:
Our multi-agent LLM architecture is highly capable but resource-intensive. We've pushed our local hardware to the absolute limit. Now, we need robust cloud servers and extended LLM API limits to deploy, test under load, and scale.
We are actively looking for external investment / smart money to cover our server infrastructure and initial marketing expenses. If you are interested in next-gen EdTech AI and want to see what we've built under the hood, let's connect!
One thing I'd validate before raising for infrastructure is where your competitive advantage actually comes from.
If it's the quality of the learning experience, more compute simply helps you scale it. If the advantage depends on having a resource-intensive architecture, I'd keep asking whether some of that intelligence can be moved into product design rather than model complexity.