Convex. Reactive backend platform with database, functions, and deployments.
Connect your Convex team to map its projects and deployments into the context graph. Agents can then reason about which environments exist, where each project's production deployment lives, and what changed between syncs.
Join the waitlistWhat Polylane does with Convex.
Projects and deployments in the graph
Every project with its production, preview, and development deployments, their regions and URLs, next to everything else you run.
Production separated from the noise
Dev and preview deployments stay visible on the map, but only production changes reach change intelligence and incidents.
Change history on every sync
New deployments, deleted projects, and re-pointed production deployments land in the change feed with the sync that found them.
Your data stays yours
Polylane reads the Convex Management API and nothing else: it never queries your tables and never holds deploy keys.
How it works.
Connect
Paste a team access token from your Convex team settings. It's encrypted before it's stored, and the agent never sees it.
Sync
Projects and deployments land in the graph, and every sync after records what changed.
Agents use the map
Investigations reach from a service to the Convex deployment behind it, with its region, class, and deploy history in hand.
In the graph.
The Convex resource types Polylane syncs, with their edges, config, and history.
Project Deployment The tools agents get.
What an agent can do with Convex mid-investigation. Every call lands in the thread transcript, cited.
convexApi Call any Convex Management API endpoint that lacks a dedicated tool
convexGetLogs Get function execution logs for a Convex deployment
convexQueryMetrics Query function metrics for a Convex deployment
Questions.
Is Polylane an AI SRE for Convex?
It's what most people mean by the term: agents that keep your Convex resources in one context graph, watch them, investigate issues and fix what they can, with a human approving anything that changes production. We call it self-operating software rather than an AI SRE: the point is that your software looks after itself, not that a bot joins the on-call rotation.
Can Polylane change anything in Convex?
Only if you let it. Connected read-only, agents investigate and explain, and changes only ever land as pull requests for you to review. With write access, every write pauses for your approval first, with the method, path and reason on screen. Rollbacks are off by default: you enable them per automation and can turn them off at any time.