Skip to content
memFrame

memFrame

memFrame brings a pandas-like DataFrame API to DuckDB, PostgreSQL, and ClickHouse — async-first, with an optional AI agent for natural-language data work.

What It Provides

  • Database-backed DataFrame API across DuckDB, PostgreSQL, and ClickHouse.
  • Each pandas-style call is compiled to backend-native SQL and executed on the engine — your data never leaves the database.
  • Async-first surface with sync equivalents for every operation.
  • Upload from CSV, Parquet, or pandas DataFrame.
  • Inspection, selection, cleaning, statistics, arithmetic, sorting, reshape, merging, Plotly charts.
  • Two-level cache: lineage audit + replayable result tables.
  • Optional AI agent layer (memframe_ai) for natural-language data work, including one-shot AI dashboards (adashboard) with a built-in query guardrail.

Start Here

  1. Read Architecture for a map of the call path and subsystems.
  2. Follow Getting Started to connect and upload data.
  3. Try the AI Agent for natural-language data work.
  4. Read Upload Manager for ingestion behavior.
  5. Read Dataset Operations to manage uploaded datasets.