Run the engine on your terms.
Start with one Rust process or separate control from bidding. Keep catalog distribution, spending authority and operational signals explicit.
Coming soon. Request early access.
Separate responsibilities as the deployment grows.
Start with one process.
The all role combines control, bidding and events, with the implemented SQLite store for local persistence.
Distribute the buying state.
Bidders fetch versioned catalog snapshots and rebuild their indexes away from the auction path.
Allocate authority and observe.
Budget leases limit local spending authority. Health checks, structured logs and metrics expose operational state.
Ship with an operating model.
Keep auctions close to memory.
The bidder reads an immutable compiled index. Catalog writers prepare the next version before swapping it into use.
Know when an instance is ready.
Liveness, readiness and version endpoints accompany Prometheus metrics and structured request logs.
Prepare for rollout and recovery.
The repository includes deployment assets, configuration validation, operations guidance and event-spool recovery behavior.
Which storage backends are implemented?
The server currently supports SQLite and an in-memory store. A Postgres backend has been added to the storage layer, with pooled connections, migrations and catalog notifications. Its server configuration integration is still in progress.
Switch to a modern,
affordable DSP.
Buy programmatically or go direct, with agentic workflows and validation included.