Founder & CTO at IgnitionRAG — the self-service RAG infrastructure platform.
I build AI backends so companies don't have to. Import documents, create AI agents, deploy widgets — in minutes, not months.
IgnitionRAG is a "backend IA administrable" — think Supabase, but for AI on enterprise documents.
- 12-parameter configurable RAG pipeline with built-in A/B testing
- REST API, TypeScript & Python SDKs, MCP server (23 tools)
- 5 embeddable widget types (chat, search, FAQ, form, webhook)
- Visual workflow builder (37 node types)
- BYOK: zero markup on LLM/embedding costs
- Dashboard for non-technical teams (import, feedback, ROI)
LangGraph, Qdrant, Jina AI, Drizzle ORM, Clerk, Stripe, MCP Protocol
AI Engineer with enterprise experience at BNP Paribas CIB (on-premise RAG for banking docs), Sanofi (RAG pipelines for pharma marketing), Brevo (GenAI platform, 50K beta testers), and Suez. Published author at Packt Publishing (Agentic AI Apps with Next.js, TypeScript & MCP).
Former fullstack consultant at WebExpr (SharePoint/React for BNP Paribas), SFEIR (Crédit Mutuel), and Anka. Ex-military.
Certifications: TensorFlow Developer, Azure AI Engineer (AI-102), AZ-900, ML Stanford (Andrew Ng), Rust LLMOps (Pragmatic Labs). ENI School graduate (2021). Based in Troyes, France.
Creator of GenAI Labs (GenAI course in French) and YouTube @salim4n.
I spent years assembling RAG pipelines for enterprise clients. Then I built the tool I wish I had.
