I’m Subhajit Mukherjee, and I’d be thrilled to share my experience and insights on “The Democratization of AI” at the Web Summit in Lisbon. As someone deeply involved in helping startups leverage innovative technologies at Reckonsys, I’m excited to discuss how AI’s most powerful tools are now accessible to even the leanest teams.

Session Overview:

In this talk, I’ll walk through practical ways startups can implement Retrieval-Augmented Generation (RAG) and agentic RAG, as well as showcase how to build an AI chatbot on AWS Bedrock in under 10 minutes—no coding required.

1. Demystifying RAG: I’ll introduce RAG as a highly effective, budget-friendly AI approach where a retrieval system pairs with large language models (LLMs) to pull in precise, real-time data. This method enables startups to build powerful AI systems that don’t require extensive training or massive datasets, making RAG a practical choice for startups looking to scale quickly.
2. Agentic RAG for Automation: Moving a step further, I’ll discuss how agentic RAG empowers systems to autonomously gather, interpret, and respond to relevant information. This approach can help startups design dynamic, responsive systems that keep users engaged and informed, without the overhead of a large technical team.
3. No-Code AI Chatbot in 10 Minutes with AWS Bedrock: I’ll also demonstrate the simplicity and speed of building a fully operational chatbot on AWS Bedrock. By walking through steps like loading documents from AWS S3, chunking data, embedding with OpenAI’s text-embedding-ada-02, and storing results in a vector database (such as Qdrant), I’ll show that it’s entirely feasible to create a robust AI tool for customer service or engagement—without a single line of code.

Conclusion:

My goal is to illustrate that the democratization of AI is not a far-off ideal but a current reality. By leveraging accessible AI frameworks and tools, startups can now compete on a level playing field with major tech companies. This session will empower founders and developers alike to see AI not as a luxury but as a vital part of their toolkit