Constructive Open Sources Agentic DB for AI Agent Memory and Knowledge
Constructive has announced in a press release the open-source release of Agentic DB, a Postgres-based database architecture built to give AI agents persistent memory, structured knowledge, and hybrid retrieval. The database is designed to replace file-based memory systems commonly used in agent frameworks.
Agentic DB ships as a single Postgres schema that can be installed in one command using the PostgreSQL Package Manager (pgpm). It includes long-term memory with vector and full-text search, conversation and tool tracking, task orchestration, and runtime observability. Each text-bearing table is indexed for semantic, statistical, fuzzy, and spatial retrieval, with an auto-embedding pipeline powered by Postgres triggers and Ollama.
The release also includes Agent Skills, a command-line interface, and a type-safe SDK automatically generated from the schema. These tools allow agents to read and write data programmatically and integrate directly with frameworks such as OpenClaw. Agentic DB supports popular AI assistants including Claude, Cursor, Copilot, and others.
Agentic DB is available under the MIT license for local and single-developer use, with a cloud version in development for teams requiring secure, scalable deployments. Developers can access the release on npm and GitHub, with more details available at constructive.io.
We hope you enjoyed this article
Consider subscribing to one of our newsletters like Daily AI Brief.
Also, consider following us on social media:
Daily AI Brief
Daily report covering major AI developments and industry news, with both top stories and complete market updates
Whitepaper
Tensordyne Napier: What If One Rack Could Do the Work of Nine?
Tensordyne
This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.
Read moreYou may also like
PhoenixAI Wins 2026 Tech Ascension Award for Enterprise Agents
pgEdge Launches Starfleet Postgres Cloud Platform
AIPOCH Releases Open-Science Research Workbench
Simplilearn and Columbia Engineering Launch Agentic AI Course
DaVinci Commerce Adds Brand Agent for AI Shopping
Daily AI Brief: the AI news that matters, in your inbox.