In a significant development for the AI infrastructure landscape, Tiger Data has unveiled Ghost, a database service meticulously crafted for AI agents. This launch addresses a critical gap in the current infrastructure, where traditional databases often fall short in supporting the unique needs of AI-driven workflows.
Revolutionizing AI Agent Collaboration
As developers increasingly deploy coding, research, and workflow agents at scale, they encounter significant challenges with existing database solutions that were not designed for agent collaboration. Ajay Kulkarni, co-founder and CEO of Tiger Data, articulated this concern: “As teams run coding agents, research agents, and workflow agents at scale, they are discovering that the infrastructure underneath them wasn’t designed for the way agents actually work. Ghost is built specifically for that environment, providing limitless terrain for agents to experiment without putting production at risk.”
Ghost offers a robust framework for experimentation, allowing agents to operate in isolated environments. This isolation ensures that when an agent encounters an error, the impact is contained within a single database, safeguarding the integrity of shared resources. With Ghost, users gain access to unlimited Postgres databases, featuring rapid forking capabilities and a variety of database options ranging from ephemeral to dedicated always-on instances.
Furthermore, Ghost introduces a cost-effective model for experimentation. Historically, the expense of creating numerous isolated Postgres databases for each task or hypothesis has deterred teams from fully leveraging their capabilities. Ghost’s innovative per-query pricing structure significantly lowers the barrier to entry, making it feasible to adopt trial-and-error methodologies at an unprecedented scale.
- Free tier offering 100 compute hours per month and 1TB of storage, along with hundreds of databases and forks.
- Seamless compatibility with any MCP-enabled agent harness, including Claude Code, Cursor, Windsurf, Codex, Gemini CLI, and VS Code.
- Robust dedicated tier for production use, starting at /month, built on the high-performance capabilities of Postgres and TimescaleDB.
Jacky Liang, Developer Relations Lead at OpenRouter, shared insights on the user experience: “I don’t use Ghost. My Claude Code agent does. If you asked my CC sessions how they feel about Ghost, it’s that it’s the easiest and most agentive-native database they’ve ever used. The CLI/MCP-only experience is genius.”
Ghost is constructed on the solid foundation of Postgres, which provides a unified query language, a consistent operational model, and a rich ecosystem of extensions. This familiarity allows teams to leverage their existing knowledge while exploring new possibilities in AI agent collaboration.
Available now, Ghost includes a free tier with generous compute and storage options, alongside controlled spending caps for paid tiers. Comprehensive documentation, deployment guides, and quickstart resources can be found at tigerdata.com.
About Tiger Data – Tiger Data, the innovative team behind TimescaleDB, specializes in developing open-source time-series databases built on PostgreSQL. They also manage Tiger Cloud, a platform designed for time-series, analytics, and AI workloads. With Ghost, Tiger Data extends its Postgres platform to meet the evolving infrastructure demands of AI agents and their development teams.
Media Contact: Erica Anderson, Offleash PR for Tiger Data.
