AI knowledge management
Every company has a wiki that was accurate in March. This job answers the internal questions, finds the information across your systems, keeps the procedures current, and flags the documentation that has gone stale. When the docs do not cover it, it says so.
Concretely
- Answers internal questions from what is actually written down.
- Finds information across the systems it is connected to, not just one wiki.
- Organizes standard operating procedures so they can be followed.
- Updates documentation as answers change.
- Identifies outdated documentation rather than leaving it to be discovered.
It never invents an answer
This is the rule the whole job depends on. When the documentation does not cover a question, it says so and brings in the person who knows. An internal knowledge tool that fills gaps with plausible guesses is worse than no tool, because a confident wrong answer about your own process gets acted on, and nobody thinks to check the source.
Documentation rots quietly
Nobody notices a stale procedure until someone follows it and something goes wrong. The rot is invisible because the document looks the same as it did when it was true. Identifying outdated documentation is a continuous job rather than an annual audit, and it is the half of knowledge management that never gets staffed.
Answering and maintaining are the same job
Treating them separately is why wikis die. The questions people ask are the map of what the documentation should cover and where it is unclear, and an answer given today is a document that should exist tomorrow. Doing both means the knowledge base grows along the lines people actually use rather than the lines someone imagined at setup.
The jobs this comes from
Sanaf holds thirty five written job descriptions. These are the ones behind this page.