Operations | Monitoring | ITSM | DevOps | Cloud

Continuous delivery is only as good as your ability to understand what’s happening inside your pipelines. When a build is slow, flaky, or burning through capacity, you need more than a green/red status and a wall of logs — you need traces. Bitbucket Pipelines now exposes pipeline execution as OpenTelemetry (OTel) traces via webhook events. This lets you stream detailed pipeline spans into your own observability stack and correlate them with the rest of your system. This post walks through.

Seven developer-requested features. Tighter control over branches, history, and large repos. No CLI detours required. If you have been using GitKraken Desktop in a complex repo, you already know what it feels like when the commit graph turns into a wall of branches. When rebasing requires more ceremony than it should. When you just need one file back from three commits ago but have to orchestrate a whole checkout to get it. GitKraken Desktop 11.10 is built for those moments.

New Australian research reveals poor website search is costing businesses revenue as AI raises the bar.

Somewhere between a model’s first demo and its first production workload, the cost conversation changes completely. Training is a big number, but it’s a finite one. Inference isn’t. Every user interaction, every query, every API call triggers compute behind the scenes — and unlike training, inference never stops billing. That shift from one-time expense to ongoing operational cost is where inference economics begins.

Years in the FinOps trenches taught me one thing: The data has never been the problem. The data exists. It’s out there, scattered across provider invoices, buried in tagging gaps, locked behind dashboards that maybe three people in your org actually know how to navigate. The real problem? Nobody can get to it when they need it. Engineers ship features without understanding what they cost the business, let alone whether they improved margin.

Everyone is adding Model Context Protocol (MCP) servers to everything right now. And I get it. MCP is clean. It’s standardized. You write a server, expose some tools, and suddenly your LLM can query your log platform, pull a dashboard, and fire an alert. It feels like the right abstraction. But I’ve watched teams at serious companies burn weeks building MCP integrations for workflows that should have been skills, and build skills for things that genuinely needed MCP.

Kentik’s AI Advisor acts as a virtual network engineer, helping teams of all skill levels troubleshoot, manage, and optimize their infrastructure with unprecedented speed and context. We explore seven practical NetOps use cases, from rapid incident triage and capacity planning to upcoming live-device command support, that demonstrate how using AI as a collaborative teammate dramatically reduces manual investigative work.

While tracing is a critical component of any observability strategy, metrics — especially RED metrics (request rate, error rate, and duration) — are widely considered the gold standard for monitoring service health. Tempo, the open source, easy-to-use, and highly scalable distributed tracing backend, is well known in the OSS community for storing and querying traces. It can also, however, generate RED metrics directly from those traces using the optional metrics-generator component.

Modern cloud environments include thousands of resources across providers, teams, and accounts. Organizations need the ability to quickly locate the right resources so that they can manage resource compliance and troubleshoot issues. When engineers need to answer questions such as which databases are still on extended support or which storage buckets lack encryption, they often have to switch consoles, use provider-specific query languages, and know obscure version strings or configuration flags.