Langfuse is for this scene: the LLM service is live and users are complaining about output quality. Open the monitoring dashboard — CPU normal, memory normal, P99 latency within SLA. From an infrastructure perspective, the service looks perfectly healthy. This is the LLM observability gap: traditional APM tools can tell if the system is healthy, but not if the AI is.

Note: Redis 8.0 reached GA in 2025, adding native vector search and more. Redis 7 features only; Redis 8 upgrade is a separate topic.

Redis 7 is more than a version bump—it unifies search, JSON support, and graph databases into one coherent platform. Key changes and when to use what follow.

Use MCP when a tool must be reused across Hosts. GitHub and filesystem pay off; high-frequency calls and transactional paths do not. Least-privilege tokens, pinned versions, per-Server timeout.

Slow SQL shows up in the plan first, not in postgresql.conf. Follow PostgreSQL: run EXPLAIN (ANALYZE, BUFFERS), see Seq Scan versus Index Scan, then choose an index, a rewrite, or a knob.

Missing indexes, a function wrapped around a column, stale statistics, autovacuum falling behind, cost constants left on the spinning-disk default — any of those send the same query through a full table. Read the plan before editing config.

Go 1.23 shipped in August 2024, bringing the long-awaited range-over-func feature. Looking back six months later, which features actually changed day-to-day development? This is a retrospective from real use, not a release announcement.

An LLM process that “looks fine” on CPU is often already serving a worse model, a slower first token, or a truncated completion. Prometheus + Grafana will show requests; a default node dashboard will not show tokens, TTFT, or a model label exploding cardinality.

The path is scrape → query → rules → panels. Every llm_* name below is an example metric name, not a standard export from Prometheus or from any particular gateway. Match /metrics before copying queries.