Operations | Monitoring | ITSM | DevOps | Cloud

Engineers have been shipping pieces of "the graph" for years. Service maps. Dependency graphs. Knowledge graphs. RDF triples. The newest entrant is the context graph, and the reason it shows up now is specific: software is increasingly executed by agents, and agents need a model of how work actually happens, not just an index of what exists.

Every so often, NVIDIA releases something that quietly changes the direction of the industry. CUDA did it. DGX did it. NVLink did it. Vera Rubin feels like one of those moments again. At first glance, Rubin looks like the natural successor to Blackwell. Faster GPUs, larger memory pools, and eye watering performance numbers. But the more you dig into the architecture, the clearer it becomes that NVIDIA is not simply shipping another accelerator generation.

Customers regularly come to us looking to solve their observability problem by connecting the dots from frontend to backend. It sounds straightforward in theory, but in practice it's one of the hardest problems in modern application monitoring. The frontend monitoring tools they already have in place tend to be proprietary or narrowly scoped to frontend needs, leaving them without the context-rich backend data that makes real triage possible.

Internal visibility isn’t enough for modern incident response. Your app team has three dashboards open and everything looks fine. CPU is healthy, memory is stable, the application servers are responding normally. But users are still complaining. The checkout page is slow. Logins are timing out. Support tickets are piling up. And your monitoring tools have nothing useful to say about why.

Today we’re launching Lightrun AI Skills, structured, repeatable investigation workflows built for AI coding agents. With Lightrun MCP, agents like Claude Code, Codex, and Cursor can already instrument live production services and reason over live runtime evidence without a redeployment. But AI agents remain non-deterministic by design, using the same tool differently every session.

When your mail server connects to a recipient server to deliver email, the very first thing it does after the TCP connection is established is introduce itself. That introduction happens through the EHLO command (or its older predecessor HELO), and it looks like this: That hostname in the EHLO line is your SMTP banner. It is what your server claims to be.

The Start of Authority record is the first record in any DNS zone file. It's the record that says "this zone exists, this is the primary nameserver in charge, and here are the timing rules that govern how this zone behaves." A full SOA record looks like this when you query it: Each of those numbers does something different. The one that triggered your warning is the Expire value, the fourth number. In this example, 1209600 seconds, which is exactly 14 days.

Kerberos is a secure network authentication protocol that allows users and systems to prove their identity over a network without sending passwords in plain text. It is widely used in enterprise environments (for example, in Windows domains) to enable single sign-on (SSO). At its core, Kerberos uses a trusted authority called the Key Distribution Center (KDC) to issue encrypted “tickets” that verify identity.

Over the weekend, my own site went down and Oh Dear didn't catch it. The origin server had fallen over, but Cloudflare happily kept serving the cached HTML. Everything looked fine from the outside. Embarrassing. Scratching our own itch here, we just shipped magic variables: short placeholders you can drop into your monitor URL, request headers, or POST payload. Right before each check, we replace them with fresh values, so every request is unique enough to slip past any cache and actually hit your origin.

Measuring AI’s impact on your engineering team is harder than it sounds. Headlines claim AI writes 30% of code and doubles productivity, but those numbers rarely match what you see on the ground. Without a dedicated dashboard that blends leading indicators, anti-gaming safeguards, and ROI reporting, you cannot answer the question that matters most: is AI helping your team ship better software faster?