AAI Developer·21h agoDon't Put the Prompt on the Span: Tracing Django AI Agents with OpenTelemetry (Part 6)Trace Django AI agents with OpenTelemetry GenAI conventions: PII-safe spans, prompt refs, tail sampling on cost, and debugging a bad answer.15 min read·aidevpro.hashnode.dev
TThyne Ventures·1d agoWhich Snowflake tasks keep failing quietly? One read-only query to checkScheduled tasks that fail every run still burn credits and leave stale data. Here's how to spot them in a minute.2 min read·thyneventures.hashnode.dev
AAiOps Community·1d agoA practical checklist for opening a technical community to AI agentsA practical guide to agent onboarding, identity, moderation, and quality controls for technical publishing communities.3 min read·aiopscommunity-cyntra.hashnode.dev
AAbdurashid Akbarov·1d agoConnect product analytics and debugging with a shared context contractA practical context contract for connecting analytics, errors, traces, and feature flags, with identity, privacy, and staging-test guidance.7 min read·abdurashid-builds.hashnode.dev
Nnaresh kumar·2d agoWhat 84,800 local spans taught me about my coding agents and my local AI servicesToken trackers tell you what your agents cost. They do not tell you why a child agent hung for ten minutes.8 min read·nareshh74.hashnode.dev
SSuprith Modulis·2d agoObservability Cost Is an Architecture Decision, Not Just a Vendor BillLearn why monitoring bills increase as applications scale and how to audit logs, traces, retention and tooling to build a cost-efficient observability stack.5 min read·modulis-engineering.hashnode.dev
UUmer Amin Durrani·2d agoQueue Depth Is a Weak Alarm on Its OwnBacklog size alone cannot tell you whether customers are waiting too long.3 min read·umeramindurrani.hashnode.dev1M
AAbhijeet Shinde·3d agoThe Slow Requests Weren’t My Code. I Just Couldn’t See the Network LocallyWhat my internship taught me about production, Grafana, and a public address that sent traffic the long way round5 min read·weeklyupdate.hashnode.dev
CCooper Email·3d agoLLM observability pack: free tracing plans for Langfuse, PostHog, LangSmith and moreOnce an LLM app or agent leaves your laptop, you need to see what it actually did: which prompts went out, what came back, how long each step took and where it failed. Almost every tracing and eval to4 min read·perq.hashnode.dev
SSamuel Aniekeme·4d agoBuilding a Check Engine Light for My Homelab with GrafanaWhy I stopped building wall art Most homelab Grafana dashboards are beautiful and useless: forty graphs, a world map of blocked IPs, and nobody knows which line should worry them. I rebuilt mine aroun26 min read·10xannie.hashnode.dev
