Meridian

Built a scalable AI workspace for interacting with documents and knowledge through retrieval-augmented generation.

Implemented document ingestion, chunking, embedding generation, vector search, and context-aware response generation.

Implemented multi-model support, allowing users to switch between LLMs and lightweight SLMs based on their requirements.

Designed the ingestion, retrieval, and inference pipeline with a focus on scalability, low latency, and reliable contextual responses.