A civic analyst reviewing urban data on screen.

Resources

Everything you need to read IntelCit results well

Practical overviews, service guides, and the questions to ask before acting on any output.

AI-generated editorial image — temporary placeholder pending licensed field photography.

Start with the practical overview

IntelCit turns a street image, video, or location into a reviewable observation and a practical next step. A person always reviews the output and chooses what to do.

Service guides

Open any service for its inputs, outputs, evidence, and known boundaries.

Reading the evidence

Every service reports observations with confidence where supported — not verdicts.

See the full evidence method

Questions to ask before acting

A short checklist your team can use on any result.

  • What exactly did the service observe, and where is the source?
  • Is this an observation, an inference, or a confirmed human decision?
  • What confidence is attached — and is it validated accuracy or an example output?
  • Which known conditions (lighting, occlusion, angle, season) could affect this result?
  • Who reviews the output before any action is taken?

Quick glossary

The short version. The full plain-language glossary lives on Science & Evidence.

Precision

Of the things flagged, how many were correct.

Recall

Of the things really there, how many were found.

Confidence vs validated accuracy

A confidence score is the model's estimate for one output, not validated accuracy.

See the full evidence method

An illustrative scenario

A generated example, using neutral IDs and sample data.

A reviewer opens Example segment EX-014, sees a litter count with an attention level, confirms it is accurate, and records a cleanup action.

Illustrative operating scenario, not a customer case study.

Frequently asked questions

Does IntelCit decide enforcement?

No. Every output is prepared for human review; a person makes the decision.

Are the numbers shown on this site real?

No. Interface previews use generated sample data, clearly labeled.

What is the CAI score?

IntelCit's internal cleanliness scale from 1 (nearly clean) to 4 (very dirty).

How do I try it with my own data?

Request a demo. The form opens your email app with a pre-filled message to info@intelcit.com.

Public documentation

Safe product concepts, a getting-started overview, a registry-derived service reference, and evidence concepts.

Open the docs

Bring a result to review together

A guided demo can walk through a service, its evidence, and the review workflow with your team.