The problem
Webhook delivery looks simple until endpoints fail, traffic spikes, or a receiver processes the same event twice. Without signing, retries, limits, and isolation, integrations become unreliable and difficult to debug.
Loading page
Preparing the content for you.
Webhooks that survive the real world.
Multi-tenant webhook delivery infrastructure that signs event payloads, routes events to matching subscriptions, retries failed deliveries with backoff and jitter, enforces rate limits, supports idempotency, and isolates failed events in a dead-letter queue.
Overview
RelayGrid is delivery infrastructure for systems that depend on webhooks. Events are matched to subscriptions, signed so receivers can verify authenticity, and delivered with retries that account for transient failures.
Multi-tenancy is treated as a first-class concern. Rate limits, idempotency, and dead-letter isolation keep one tenant's failure from becoming another tenant's outage.
Webhook delivery looks simple until endpoints fail, traffic spikes, or a receiver processes the same event twice. Without signing, retries, limits, and isolation, integrations become unreliable and difficult to debug.
RelayGrid provides signed, idempotent, multi-tenant delivery with bounded retries, per-tenant rate limits, and a dead-letter queue for events that cannot be delivered.
Architecture
A high-level conceptual view of the main stages in this system.
Events enter through an API and are validated before routing.
Events are matched to the subscriptions that should receive them.
Payloads are signed and delivered with idempotency keys.
Failures retry with backoff and jitter before moving to a dead-letter queue.
Capabilities that are documented as part of the current implementation.
Receivers can verify that a payload came from RelayGrid.
Events are delivered only to matching subscriptions.
Transient failures are retried with bounded backoff and jitter.
Traffic is bounded so one tenant cannot exhaust shared capacity.
Duplicate deliveries can be detected by receivers using a stable key.
Undeliverable events are isolated for inspection and replay.
These items are planned and are not available yet. They are listed separately so completed functionality is never confused with future work.
Visibility into delivery rates, failures, and tenant health.
Controlled replay of dead-lettered events.
Published throughput and latency results under realistic traffic.
Technology
The tools and technologies associated with the implementation.
High-concurrency delivery workers and API.
Subscriptions, delivery state, and audit records.
Rate limiting and fast delivery state.
Durable event transport between services.
Operator-facing management interface.
Delivery and failure metrics.
Repeatable service packaging.
Scaling and isolation for delivery workers.
Payload signing for receiver verification.
Challenges
The engineering constraints that shaped the design.
Retries make duplicate delivery possible, so idempotency must be designed into the contract rather than assumed.
A failing or abusive tenant must not consume the capacity needed by everyone else.
Spikes in event volume require bounded queues and predictable degradation.
Outcomes
Qualitative outcomes from the design. Measured results are only reported where they have been formally evaluated.
Links
Explore the project through the destinations that exist. Missing destinations are labelled instead of being replaced with a placeholder URL.
No public link has been provided for this project yet.
No public link has been provided for this project yet.
No public link has been provided for this project yet.
No public link has been provided for this project yet.
0 of 4 external links are currently available. Unavailable links are shown deliberately rather than pointing to a placeholder destination.
Keep exploring
Browse the full portfolio or start a conversation about this project.