Open to new projects
Work / Product I started / ErrandGo
Case study · Product I started

ErrandGo.

Errands and deliveries done by people nearby, paid safely by M-Pesa.

RoleEngineer, end to end · idea shared with collaborators
ForCustomers and runners in Nairobi
Timeline2026 – ongoing
StackLaravel, React Native, TypeScript, M-Pesa
StatusIn development
227automated tests
1app for customers and runners
The problem

Busy people in Nairobi lose hours to small errands. People who could run them want flexible income. Neither side trusts the other with money.

What I built

One app for customers and runners.

Customers post an errand, runners accept it, and M-Pesa holds the payment until the job is done. An admin console handles disputes and payouts.

Key decisions

Choices that made it hold up.

  • ETAs learned from real trips

    Travel times come from past trips, with OSRM and Google Maps as fallbacks.Why: map estimates ignore Nairobi traffic

  • Public IDs, not numbers

    Every record is referenced by an ID that can't be guessed.Why: stops people walking through other users' data

  • Tests for slow queries

    Tests fail if a page runs more database queries than it should.Why: catches slowdowns before users feel them

What's next

Push notifications, a live map view and a first test on real devices.

Contact

Have an idea or a system to build?

Tell me what you need. I'll reply within a day with questions, a rough approach, or a time to talk.

Phone+254745600377
Based inNairobi, Kenya · working worldwide
What do you need?