Patroni HA tuning: Introducing the first agentic AI tuning solution for highly-available Postgres Patroni clusters. Optimize self-hosted high availability without compromising uptime, slashing infrastructure costs and latency. Read more below.
AWS Quick Launch: Deploy the DBtune agent for your RDS and Aurora instances with CloudFormation by automating EC2 provisioning, IAM configuration, etc. It is the click-click method that you have all asked for and it runs in three minutes! Learn more here.
Agent health checks: Get real-time feedback on your agent’s status with targeted debugging guides to resolve agent misconfiguration issues.
Tuning summary report: Compare your recommended configurations against your baseline with expanded performance indicators, providing the deep transparency needed to validate every optimization.
Now you can have high performance WITH high availability
The compromise ends now – DBtune breaks this trade-off with the first agentic AI tuning specifically engineered for Patroni. By integrating directly with the Patroni REST API and Distributed Configuration Stores (DCS) like Etcd or ZooKeeper, DBtune becomes "cluster-aware"—automatically identifying the active leader and propagating optimized settings across your entire topology without manual intervention.
What happens when a database professional with three decades of experience—spanning Oracle, SQL Server, and PostgreSQL—puts AI tuning to the test?
After three decades of manual tuning in high-stakes financial services environments, Frans Verduyn Lunel explored how DBtune handles the complexities of PostgreSQL optimization on everything from a Raspberry Pi to a MacBook Pro.
The highlights:
"Fluitje van een cent": Getting started with DBtune, adding the database, and setting up the agent was “a piece of cake setup.”
Precision: Automated tuning, e.g., work_mem, based on real-time workload fingerprints.
Control: Leveraging performance guardrails to filter out workload anomalies.
"DBtune is remarkably easy to work with. My impression is that DBtune achieves better results than I could manually." he concludes.
Ready to shift from reactive performance troubleshooting to automated tuning maintenance? Read Frans' blogon Medium.
If you’re heading to Amsterdam for KubeCon + CloudNativeCon Europe 2026, this should be on your to-do list. DBtune is sponsoring a bowling lane at KuBBBecon – BeNeLux Bowling Borrel by HCS and Kangaroot.
Join Farshad Poye and Mohsin Ejaz for a fun evening of bowling, networking, and cloud-native conversations — just 7 minutes from the main venue. And yes, there is a LEGO Star Wars to win for the one who knows how to throw the strikes.
POSETTE, online (17 June): Building safety tooling for risk-free AI tuning of Postgres: Fast cars need fast brakes (Mohsin Ejaz)
Articles we love
The AI inversion by Tomas Vondra In his blog, he explores how AI shifts database work from writing code to verifying it. This "inversion" fundamentally redefines the developer's role, moving from manual implementation to high-level strategic oversight and validation..
Lessons learned writing an MCP server for PostgreSQL by Dave Page Discover how Model Context Protocol (MCP) enables AI agents to interact directly with PostgreSQL, shifting the focus from manual querying to secure, automated metadata inspection and schema-aware data retrieval.
The new Postgres AI ecosystem by Shaun Thomas This presentation explores the evolution of PostgreSQL into an intelligent data hub, highlighting pgEdge's AI stack, MCP servers, and vector tools that enable seamless, natural language interactions with your database.
Join our team
We are hiring a UX Designer to bridge the gap between our complex AI-driven optimization and the humans who use it:
Lead the end-to-end design of DBtune's dashboard—translating intricate PostgreSQL performance metrics into clear, actionable visualizations for environments like AWS RDS and Azure Flexible Server.
Collaborate with research and engineering teams to simplify AI interactions, maintain our design system, and conduct user research to ensure our platform remains intuitive and transparent.