Technical Infrastructure
how to set up and run your ai solution technically
This card covers how your AI tool is set up and run on the technical side. Where it lives, what it connects to, and who keeps it working all affect reliability and cost. For a newsroom with limited engineering time, a simple setup you can maintain often beats a powerful one you cannot.
Questions to explore
- Where will the tool run, and who has the skills to keep that running?
- What other systems does it need to connect to, and how stable are those links?
- How much engineering support does this realistically need, week to week?
- What happens to the newsroom's work if the tool goes down during a busy day?
- How does the setup handle sensitive data, both at rest and in transit?
Expert voices
“Weigh cloud, local, and hybrid setups against the sensitivity of your content. Local AI models may be necessary to analyze confidential or surveilled documents.”
“Decide deliberately when a tool should run locally rather than in the cloud. Where the processing happens is an editorial choice, not just a technical one.”
Things to consider
- Match the setup to the technical help you actually have, not what you wish you had.
- A tool no one can maintain becomes a liability over time.
- Plan for what happens when a part of the system fails.
Pull Technical Infrastructure when it is relevant and set it aside when it is not. Pair it with the other AI Solutions cards, lay them out on a table, and use the questions above to get everyone on the same page. Capture what you discuss on sticky notes or in a shared doc.
More AI Solutions cards





