What is Neum AI
Neum AI provides a RAG-first framework designed to build performant, scalable, and reliable data pipelines. It focuses on key data transformations such as loading, chunking, and embedding. The platform offers built-in connectors to common services, allowing users to choose from various data sources, embedding models, and vector databases. Additionally, users can add their own connectors using the open-source framework. The platform supports testing and deploying pipelines locally using open-source SDKs and directly deploying them to the Neum AI cloud.
How to Use Neum AI
- Use the open-source SDKs to compose and test your data pipelines locally.
- Choose from built-in connectors for data sources, embedding models, and vector databases.
- Deploy your tested pipelines directly to the Neum AI cloud for production use.
Features of Neum AI
-
Open-source SDKs
Compose data flows for building performant, scalable, and reliable RAG pipelines.
-
Built-in connectors
Choose from connectors for data sources, embedding models, and vector databases, or add your own.
-
Test and deploy pipelines
Run pipelines locally using open-source SDKs and deploy them to the Neum AI cloud.
-
Scale
Distributed architecture optimized for embedding generation and ingestion for billions of data points.
-
Sync
Keep vectors in sync with built-in pipeline scheduling and real-time syncing.
-
Observability
Monitor data to ensure correct syncing into your vector database.
-
Smart Retrieval
Built-in retrieval informed by the organization of your data and associated metadata.
-
Self-improving
Improve context quality by providing feedback on retrieval quality.
-
Governance
Observe actions like searches and data movements.