Projects That Get You Hired: 8 African Stack App Ideas

Every junior developer's portfolio looks the same. A to-do app. A weather app. An e-commerce clone that accepts fake credit card numbers. These projects prove you can follow a tutorial. They do not prove you can build something useful in the market you are trying to work in.

Kenyan tech companies do not accept credit cards. They accept M-Pesa. They do not send push notifications to iOS apps. They send WhatsApp messages. They do not assume all users have smartphones. They build USSD fallbacks. When a hiring manager sees a portfolio full of generic projects, they see a developer who learned to code but does not understand the market.

When they see a project that integrates the Daraja API, handles M-Pesa callbacks, sends WhatsApp confirmations, and falls back to SMS for feature phone users? That developer understands what it takes to ship a product in Kenya. That is the developer who gets the interview.

The projects below are designed with this in mind. Each one solves a real Kenyan problem using real Kenyan infrastructure. They are scoped to be buildable in 2 to 4 weeks, not as semester-long endeavors. And each one teaches skills that directly transfer to the work Kenyan tech companies do every day.

The problem: Millions of Kenyans belong to chamas (informal savings groups). Most track contributions in exercise books or WhatsApp group messages. Money gets lost. Records are disputed. The treasurer's notebook is the single point of failure.

What you build: A web app where chama members can see the group balance, contribution history, and upcoming payment dates. Contributions are collected via M-Pesa (STK Push from the app). Every transaction is recorded in a database. The treasurer gets an admin dashboard. Members get SMS or WhatsApp notifications when contributions are received.

Skills demonstrated: Full-stack web development, M-Pesa Daraja API (STK Push and callbacks), user authentication by phone number, transaction history and reporting, WhatsApp or SMS notification integration.

Scope tip: Start with a single chama (no multi-tenancy). Focus on contributions first, ignore loans and penalties for the MVP. Add those features later if you want to expand the project.

The problem: Small restaurants and food vendors in Nairobi take orders over WhatsApp manually. The owner types the same menu 30 times a day, calculates totals by hand, and hopes customers actually pay via M-Pesa before delivery.

What you build: A WhatsApp bot using the Cloud API that shows the restaurant's menu (using interactive list messages), takes orders, calculates totals, triggers M-Pesa STK Push for payment, and sends order confirmation. An admin dashboard shows incoming orders, payment status, and daily revenue. Orders only go to the kitchen after payment is confirmed.

Skills demonstrated: WhatsApp Business API integration, webhook handling, conversation state management, M-Pesa payment processing, real-time order management.

Scope tip: Start with a fixed menu (no customization). Skip delivery logistics for the MVP. Focus on the ordering and payment flow being smooth. A working demo with 5 menu items that actually processes sandbox M-Pesa payments is more impressive than a pretty UI with fake checkout.

The problem: Small lending organizations and SACCOs need to let members check balances and make repayments from any phone, including feature phones.

What you build: A USSD application (via Africa's Talking) that lets members dial a code to check their loan balance, view repayment schedule, and initiate an M-Pesa repayment. An admin web portal shows all member accounts, loan status, and repayment history.

Skills demonstrated: USSD session handling and menu design, integration with Africa's Talking API, M-Pesa payment integration, database design for financial records, admin dashboard development.

Scope tip: Use Africa's Talking sandbox. Do not worry about getting a real shortcode. Focus on the session flow working correctly: member dials code, identifies themselves (by phone number), sees their balance, initiates repayment, gets SMS confirmation. The web admin panel can be basic. Clean data display matters more than fancy charts.

The problem: Boda-boda riders in residential estates get customers through word of mouth and random street hailing. Customers have no way to request a specific rider or book in advance. Riders have downtime between jobs.

What you build: A simple web app (mobile-first) where customers in a specific area can request a boda ride. Available riders see incoming requests and accept them. The customer gets the rider's name and phone number. After the ride, the customer pays via M-Pesa (pay the rider directly or through the platform with a small commission). Both parties can rate each other.

Skills demonstrated: Mobile-first responsive design, real-time features (rider availability, request matching), M-Pesa integration for payments, user rating system, geolocation basics.

Scope tip: Do not try to build Uber. Focus on one estate or neighborhood. Skip real-time GPS tracking (that is complex). Focus on the request-accept-pay-rate cycle. The value proposition is "I can call my trusted boda guy through the app" not "I can track a random rider in real time."

5. School Fee Payment Tracker. Parents pay school fees via M-Pesa. The school needs to know who has paid, how much, and what balance remains. Build a system that receives M-Pesa payments (using paybill or buy goods), matches them to student accounts, and sends parents an SMS receipt. Admin dashboard for the school bursar showing payment status by class and student. This solves a real pain point for every private school in Kenya.

6. Tenant Rent Collection System. Landlords managing 10 to 50 rental units collect rent via M-Pesa and track it in notebooks or spreadsheets. Build a system with M-Pesa payment tracking (auto-match payments to tenants by phone number or account reference), automated SMS reminders for overdue rent, and a landlord dashboard showing occupancy, payments, and arrears. This is a product you could actually sell.

7. Agricultural Marketplace with USSD Access. Farmers in a county want to list their produce and connect with buyers. Build a web platform where buyers browse and order, plus a USSD interface where farmers can list produce, check prices, and confirm sales using any phone. Payments via M-Pesa. SMS notifications for both parties. This demonstrates multi-channel delivery (web + USSD) and agricultural domain knowledge.

8. Event Ticketing with M-Pesa. Build a simple event listing and ticketing platform where organizers create events and attendees buy tickets via M-Pesa. After payment, the system sends a WhatsApp message with a QR code ticket. At the event, the organizer scans QR codes to validate tickets. This combines web development, M-Pesa payments, QR code generation, WhatsApp delivery, and a simple validation system. Scope it to one event at a time for the MVP.

Building the project is half the work. Presenting it is the other half. Here is how to make your African Stack projects stand out in a portfolio.

Write a proper README. Explain the business problem in plain language. Describe your technical decisions (why you chose this database, how you handle M-Pesa callbacks, what happens when a payment fails). Include setup instructions that actually work. An employer who can clone your repo and run your project locally in 5 minutes is already impressed.

Record a demo video. A 2-minute video showing the user flow (customer places order on WhatsApp, M-Pesa prompt appears, payment confirms, receipt is sent) is more convincing than any code review. Use your phone to show the WhatsApp side and screen recording for the admin dashboard.

Show the hard parts. Anyone can build a CRUD app. Employers want to see how you handle edge cases: what happens when an M-Pesa payment fails? What if the user's USSD session times out mid-transaction? What if two customers try to book the last available slot simultaneously? Document how you solved these problems.

Deploy it. A project running on a live URL (even a free tier) shows you understand deployment. A GitHub repo with only local setup instructions shows you understand development but not operations. Deploy the web app. Keep the sandbox integrations working so the M-Pesa and WhatsApp flows are demonstrable.

If you want to build projects like these with mentor guidance and structured feedback, our Full-Stack Software and AI Engineering course (KES 120,000) has you shipping 8 production-grade projects over 26 weeks. African Stack integration is core curriculum, not an elective.