Skip to main content
All example use casesIllustrative example

Building damage from drone and satellite imagery

A shelter NGO wants to classify building damage from imagery so field teams visit the worst-hit areas first.

The challenge

A shelter-rehabilitation NGO has more damaged buildings than assessors. The team wants to classify building damage from drone and satellite imagery to decide where to send field teams first. Field staff will verify every result.

Kind of support

  • Feasibility check
  • Open datasets and models
  • Workflow with field verification

How the mentorship could run

  1. Weeks 1 to 3: check what imagery is available, how recent it is, and whether its resolution shows the damage the team needs to see.
  2. Weeks 4 to 6: test open damage-assessment datasets and models on a sample area that field teams have already visited.
  3. Weeks 7 to 9: build a workflow that ranks areas by likely damage and sends the list to field teams for verification.
  4. Weeks 10 to 12: compare the ranking with field results, record where it was wrong, and decide whether it is accurate enough to use.

What to have ready before you apply

  • The areas you work in and the imagery you can access
  • Results from past field assessments, to test against
  • How field teams currently decide where to go first
  • Your rules for handling location data

Risks to plan for

  • Building locations and damage data can put people at risk in a conflict. Store them securely and share them only with those who need them.
  • Models trained elsewhere may not work well on local building types. Test against field results before relying on any ranking.
  • Imagery can be weeks old. Show the date of the imagery next to each result.

What a good result looks like

Field teams reach the worst-hit areas sooner, every result is verified on the ground, and the team knows how far it can trust the imagery.

More example use cases

  • Country: Bangladesh (Cox's Bazar)

    Sector: WASH

    Water-point dashboards with breakdown alerts

    A WASH NGO wants automated dashboards and alerts when a tap stand or latrine block stops working across 30 camps.

    Time: 1 to 2 hours per week, 8 weeksIllustrative example
  • Country: Colombia

    Sector: Cash assistance

    Finding duplicates across three cash programmes

    An INGO wants to find duplicate and likely wrong records across three programme lists, without building a surveillance tool.

    Time: 2 hours per week, 6 weeksIllustrative example
  • Country: Jordan

    Sector: Education

    Arabic reading practice with teacher review

    A learning centre wants an LLM to draft Arabic reading exercises, with a teacher checking everything before a child sees it.

    Time: 2 to 3 hours per week, 10 weeksIllustrative example
  • Country: Nepal

    Sector: Information management

    A safe assistant over SOPs and guidance notes

    An information management officer wants to build a retrieval assistant over their organisation's SOPs, safely and on a small budget.

    Time: 1 hour per week, 8 weeksIllustrative example

Know more

Want to know more before you apply? Write to contact@theaiexpertgroup.com and a member of the Mentor Council will reply.