Skip to main content

Making dashboards more usable for LLMs

I’ve been looking into agentic workflows to act as an operations assistant for the SaaS I'm working on. A big part of that work is getting the assistant to make sense of all the alert and monitoring data that pours in every day.

Passing a bunch of raw time-series data to an LLM generally doesn’t work that well. You need to tell the LLM to aggregate the data and give it the means to do so.

Using aggregates will often lead to better insights from the LLM. This is a well-known fact to anyone who has tinkered with this (at least at the time of writing this).

Humans, of course, like to build visualizations and dashboards to solve this issue (yes, yes, dashboards are often useless, but complaining about that is another blog post). LLMs can analyze them as well and in fact are pretty good at that, so the aggregate can be something both humans and LLMs can digest.

I’ve been tinkering with the idea of appending some LLM-only content to a dashboard—for example, additional context, specific details, or even some do’s and don’ts for the analysis.

Humans can ignore that stuff, but LLMs can use it to yield better results.

Additional context for the LLM added to the bottom of the chart

I’m not really sure if this is a good idea, though. You should, of course, just include this information in the prompts.

But what if you have a lot of different graphs and no clear "just analyze this" prompt that applies to all?

Also, adding it directly to the dashboard (and therefore to the screenshot the agent will capture via a tool call) can decouple the analysis agents containing the prompts from the dashboards, which may prove useful in some organizations where the infrastructure or "dashboard-builder" team is too slow to react to the AI agent team, or vice versa.

 

Comments

Popular posts from this blog

I'm not a passionate developer

A family friend of mine is an airlane pilot. A dream job for most, right? As a child, I certainly thought so. Now that I can have grown-up talks with him, I have discovered a more accurate description of his profession. He says that the truth about the job is that it is boring. To me, that is not that surprising. Airplanes are cool and all, but when you are in the middle of the Atlantic sitting next to the colleague you have been talking to past five years, how stimulating can that be? When he says the job is boring, it is not a bad kind of boring. It is a very specific boring. The "boring" you would want as a passenger. Uneventful.  Yet, he loves his job. According to him, an experienced pilot is most pleased when each and every tiny thing in the flight plan - goes according to plan. Passengers in the cabin of an expert pilot sit in the comfort of not even noticing who is flying. As someone employed in a field where being boring is not exactly in high demand, this sounds pro...

Emit structured Postgres data change events with wal2json

A common thing I see in an enterprise system is that when an end-user does some action, say add a user, the underlying web of subsystems adds the user to multiple databases in separate transactions. Each of these transactions may happen in varying order and, even worse, can fail, leaving the system in an inconsistent state. A better way could be to write the user data to some main database and then other subsystems like search indexes, pull/push the data to other interested parties, thus eliminating the need for multiple end-user originating boundary transactions. That's the theory part; how about a technical solution. The idea of this post came from the koodia pinnan alla podcast about event-driven systems and CDC . One of the discussion topics in the show is emitting events from Postgres transaction logs.  I built an utterly simple change emitter and reader using Postgres with the wal2json transaction decoding plugin and a custom go event parser. I'll stick to the boring ...

Extracting object properties from an IFC file with IfcOpenShell

Besides the object geometry information, IFC files may contain properties for the IFC objects. The properties can be, for example, some predefined dimension information such as an object volume or a choice of material. Some of the properties are predefined in the IFC standards, but custom ones can be added. IFC files can be massive and resource-intensive to process, so in some cases, it helps to separate the object properties from the geometry data. IfcOpenShell  is a toolset for processing IFC files. It is written mostly in C++ but also provides a Python interface. To read an IFC file >>> ifc_file = ifcopenshell.open("model.ifc") Fetch all objects of type IfcSlab >>> slab = ifc_file.by_type("IfcSlab")[1] Get the list of properties >>> slab.IsDefinedBy (#145075=IfcRelDefinesByType('2_fok0__fAcBZmMlQcYwie',#1,$,$,(#27,#59),#145074), #145140=IfcRelDefinesByProperties('3U2LyORgXC2f_hWf6I16C1',#1,$,$,(#27,#59),#145141), #145142...