Zep AI provides a foundational memory layer that enables the creation of personalized AI agents by integrating user interactions and business data into a Knowledge Graph. This technology allows for rapid memory retrieval and temporal reasoning, ensuring agents maintain accurate context and adapt to changing information with minimal code adjustments.
Funding
$2.3M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
AI agents often struggle to maintain accurate context and adapt to changing information due to limitations in their memory capabilities. Integrating user interactions and business data to create personalized experiences requires significant coding effort and can be slow and inflexible.
Solution
Zep AI provides a foundational memory layer that enables the creation of personalized AI agents by integrating user interactions and business data into a Knowledge Graph. This technology allows for rapid memory retrieval and temporal reasoning, ensuring agents maintain accurate context and adapt to changing information with minimal code adjustments. Zep intelligently fuses data on a Knowledge Graph, allowing developers to provide their agents with accurate and relevant user facts through a single API call. The platform marks superseded facts as invalid and identifies and stores temporal context on the graph, enabling agents to reason with changing user state.
Target Audience
Zep AI targets AI developers and businesses looking to build personalized and context-aware AI agents for various applications.
Features
- Knowledge Graph-based memory storage for efficient data fusion and retrieval
- Temporal reasoning capabilities to track and utilize changing user states
- ChatHistory class for simple integration with LangChain Expression Language or LangGraph apps
- Support for Python, TypeScript, and Go development environments
- Instant memory retrieval for relevant results in milliseconds
- Granular memory controls for tailoring memory extraction and relevance
- SOC 2 Type II compliance with privacy controls to support CCPA and GDPR
- Structured output extraction from chat history with built-in types for datetimes, emails, and more
- Dialogue classifier for instant conversation state classification