Windmill
Windmill absorbed dbt, un-gated its warehouse languages, and turned AI authoring on by default.
A side-by-side editorial comparison of Resend and NetBox — release velocity, themes, recent moves, and the top alternatives to consider.
Resend spent a month making itself something agents and third-party apps can log into.
Resend is shipping weekly, and the last six weeks split cleanly into two tracks. One is dashboard housekeeping for teams sending at volume: template folders, suppression management from both the dashboard and the API, Message-ID on webhooks and email endpoints for threading. The other is access: a hosted MCP server with OAuth built in, a one-click Claude connector, a Codex marketplace plugin bundling the MCP server and Resend's own skills, and OAuth 2.1 with PKCE so outside developers can build authenticated Resend experiences.
NetBox is spending 4.6 paying down query cost, not adding models.
The 4.6 line opened in May with genuinely new modelling — virtual machine types, cable bundles, a flat rack-group axis, ETag support on the REST API — and every patch release since has been maintenance in the same shape: a handful of enhancements, a block of performance work, a longer block of bug fixes. The performance entries are consistently about query count: N+1 elimination in GraphQL, prefetch hints, serializer caching, chunked bulk updates, a GIN index on cable paths.
Resend is shipping weekly, and the last six weeks split cleanly into two tracks. One is dashboard housekeeping for teams sending at volume: template folders, suppression management from both the dashboard and the API, Message-ID on webhooks and email endpoints for threading. The other is access: a hosted MCP server with OAuth built in, a one-click Claude connector, a Codex marketplace plugin bundling the MCP server and Resend's own skills, and OAuth 2.1 with PKCE so outside developers can build authenticated Resend experiences.
The second track is the one that changes what Resend is. Shipping an OAuth provider alongside agent-facing distribution turns an API you hold a key for into a platform other software signs into on a user's behalf — the MCP server, the Claude connector and the Codex plugin are all downstream of that auth work. The GitHub secret scanner, which detects and auto-revokes exposed keys, reads as the other half of the same shift: if keys are going to be handled by agents and third-party apps, leaked-key cleanup has to be automatic.
Expect the agent surfaces to deepen rather than multiply — scoped OAuth permissions and per-app access controls are the natural follow-on once third parties are authenticating into accounts. More marketplace listings are cheap by comparison and likely to continue alongside.
The 4.6 line opened in May with genuinely new modelling — virtual machine types, cable bundles, a flat rack-group axis, ETag support on the REST API — and every patch release since has been maintenance in the same shape: a handful of enhancements, a block of performance work, a longer block of bug fixes. The performance entries are consistently about query count: N+1 elimination in GraphQL, prefetch hints, serializer caching, chunked bulk updates, a GIN index on cable paths.
This is a project optimising for large installations rather than new use cases. GraphQL in particular is being hardened release by release — depth limiting, prefetch hints, schema corrections — which reads as API consumers hitting limits at scale. Security fixes arrive quietly inside the same lists, including audit-trail immutability and constant-time token comparison.
Expect the 4.6.x train to continue on its roughly two-week cadence with the same enhancement/performance/fix mix, and the next feature-bearing work to wait for 4.7 rather than appearing in a patch.
Other Infra & APIs products tracked by Sparkpulse, ranked by recent ship velocity. Each card links to a full editorial trajectory and lets you pivot into a head-to-head comparison with either Resend or NetBox.
Windmill absorbed dbt, un-gated its warehouse languages, and turned AI authoring on by default.
Nine releases in four days, all of them small — Casdoor is grinding on tenant-level governance.
Prowler turned its assistant from a read-only explainer into something that can change your tenant.
Depot is no longer a build accelerator — it is assembling the whole CI stack underneath itself.
Buildkite is rebuilding its MCP server around what an agent actually does with a failed build.
The self-hosted dashboard finally let you drag things around instead of editing YAML.
See all Resend alternatives → · See all NetBox alternatives →
Latest ship moves from both products, interleaved chronologically. ⚡ = editorial spark.
They serve adjacent needs but don't currently overlap on shipped themes. Resend is currently shipping more aggressively (velocity 6.3 vs 5.0), with 1 editorial sparks in the last 30 days against 0. See the at-a-glance table above for a side-by-side breakdown of velocity, recent sparks, and editorial themes.
Sparkpulse doesn't pick a winner — we score release velocity, not feature parity. Resend is currently shipping more aggressively (velocity 6.3 vs 5.0), with 1 editorial sparks in the last 30 days against 0. For your specific use case, the alternatives sections above list other Infra & APIs products to evaluate alongside.
Top Resend alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "Resend alternatives" section above for the current picks, or visit /alternatives/resend for the full list with editorial commentary on each.
Top NetBox alternatives in Infra & APIs are ranked by recent ship velocity. Browse the "NetBox alternatives" section above for the current picks, or visit /alternatives/netbox for the full list with editorial commentary on each.