AI Lab
In ProgressAutomation2026
Repetitive Task Mapper
A small automation study that maps where repetitive manual steps appear in a workflow before any code is written.
Hypothesis
Most automation value comes from a handful of steps, and mapping them first prevents over-engineering.
Problem
Automation projects often start with tooling instead of with the actual bottleneck.
Experiment design
- Log each manual step and its frequency
- Score steps by time cost and error risk
- Automate only the top-scoring steps
- Measure before and after
Technology and technique
Clustering of task descriptions with an LLM summary of each cluster.
TypeScriptLLM API
Human involvement
A person reviews and approves the model's output before it is used. The AI drafts; the reviewer decides.
Result
A short list of high-value automation candidates from a noisy task log.
Limitations
- Clusters depend heavily on how consistently steps are described.
Ethics and privacy
Workflow logs are synthetic and describe tasks, not people.
Results are a planning aid, not a performance measurement of individuals.
Next step
Turn the scoring model into a reusable worksheet.