May 2026 focused on strengthening observability for the DataDog/integrations-core Supabase integration by delivering Supavisor pooler metrics and enabling proactive alerting for pooler exhaustion in serverless environments. Implemented ~24 new Prometheus metrics exposed via the Supavisor PromEx Tenant plugin (e.g., active client/server connections, pool checkout latency histograms, client/query performance metrics) sourced from the customer Metrics API. This work closes an ingestion gap where these metrics were previously not allowedlisted and dropped by the scraper, enabling reliable dashboards and alerts. The change set included a changelog fragment rename tied to PR #23749 (commit 74e33715ab24701ba1ce638ad063b7b73eda989d). Impact: improved operational visibility, faster incident detection, and reduced risk of capacity-related outages in ephemeral compute stacks. Tech stack: Prometheus, PromEx, Supavisor integration, serverless architectures, and change management with documentation.