TIMPs (Trustworthy Interactive Memory Partner System) is an open-source AI agent with persistent memory. It stores facts, preferences, goals, and reflections from your conversations in PostgreSQL and Qdrant, then retrieves relevant context automatically in future sessions.

TIMPs features persistent memory with a 14-field schema storing facts, goals, preferences, and reflections with full audit trail. It offers dual-mode search with SQL keyword matching (ILIKE) and Qdrant vector similarity for precise retrieval. The system provides project isolation where memories are scoped per project to prevent cross-contamination between codebases or clients. It includes ephemeral mode for private sessions that leave zero persistent trace and is model agnostic supporting OpenAI GPT-4, Google Gemini, or local Ollama with provider switching capability. TIMPs offers multiple interfaces including REST API, TUI, and Web interfaces.

After each turn, TIMPs reflects on the conversation and extracts facts with an importance score (1–5). Facts are stored with 14 metadata fields in PostgreSQL and as vector embeddings in Qdrant. Future queries trigger dual-mode retrieval: SQL keyword matching + cosine vector similarity. The architecture includes a Core Agent with system prompt and tool definitions, Planner to break goals into steps, Executor to run tools (file, web, search), and Reflection to extract and score memories post-turn.

TIMPs enables AI agents to maintain context across sessions and provides precise memory retrieval capabilities. It supports various use cases including development workflows, client projects, and sensitive conversations through its ephemeral mode.

The product is designed for developers who care about their tools actually remembering context. It integrates with PostgreSQL and Qdrant databases and supports multiple AI model providers. TIMPs can be used from terminal, desktop, or browser interfaces with full command functionality available across all platforms.