Tarek Makaila

Waterflai AI - Ship customer-facing AI agents, Skip the backlog

by
Waterflai combines the ease of no-code with the power of a custom engineering stack. Our AI Architect Agent instantly turns natural language prompts into complex agent workflows, automating the tedious setup. We handle the complete RAG lifecycle—insertion, sync, and retrieval—while you retain data governance by bringing your own vector store. Deploy via widget, API, or internal chat, then track everything with real-time logs and analytics.

Add a comment

Replies

Best
Tarek Makaila
Maker
📌
Hi Product Hunt! 👋 I’m Tarek, Founder of Waterflai. Waterflai wasn't built in a vacuum. It was built from frustration. I started my career in Data science, building recommendation engines, analyzing data and navigating the nightmares of GDPR. I learned firsthand that 80% of "AI work" is usually just plumbing—cleaning pipelines instead of creating value. Later, moving into Product Management, I sat on the other side. I had ideas for AI features, but every initiative hit the "Engineering Wall." The backend always took months to catch up to the vision. I founded Waterflai to bridge that specific gap. We flip the equation: - Visual Engineering: Don't write code. Build complex pipelines visually or let our AI Architect generate them for you instantly from natural language. - Sovereignty First: We use a Zero Copy approach. We connect directly to your existing vector stores (Pinecone, Milvus). We manage the RAG logic, but the data stays yours. - Ship Fast: Whether it's a Web Component or Headless API, you deploy in minutes, not months. Proudly incubated at Station F (Paris), we are here to ensure your best ideas don't die in the backlog. By the way, we'r thinking about open sourcing the project, tell us what you think about it :)
Cesurhan Uygun

Nice approach to the RAG lifecycle problem, that's the part everyone underestimates. we built something similar at TalkBuildr but focused on agencies white-labeling chatbots for their clients. How do you handle knowledge base updates when the source content changes, do you do incremental syncs or full re-index?