← Back to selected work

Case studies

Structured drafts for the stories behind the work. Each section is ready for final evidence, visuals, and narrative when available.

01 / CASE STUDY FRAMEWORK

Metups Zimbabwe

A local second-hand marketplace designed around the realities of mobile commerce in Zimbabwe.

RoleFounder · Solo developer
StackHTML, CSS, JavaScript, Supabase
FocusTrust, discovery, mobile use

Context & challenge

Add the market insight, user problem, and the constraint that made this worth solving.

Goal & success signals

Define the intended outcome and the measures that demonstrate progress.

Approach

Explain research, product decisions, build phases, and your reasoning.

Key decisions

Show 2–3 decisions with trade-offs: local payments, moderation, listing flow, or PWA choices.

Outcome & next steps

Add launch evidence, feedback, metrics, what changed for users, and what you would improve next.

Suggested evidence: marketplace screens, the listing journey, architecture diagram, launch metrics, and a short reflection.

02 / CASE STUDY FRAMEWORK

MDUMENI AI Agronomist

AI crop guidance and market intelligence for Zimbabwean smallholder farmers.

RoleLead developer · Team of 5
StackReact Native, FastAPI, Supabase
FocusPractical agricultural guidance

Context & challenge

Describe the farming decisions users face, their information gaps, and access constraints.

Goal & success signals

Clarify the user outcome, team objective, and how usefulness or trust is measured.

Approach

Outline discovery, team collaboration, technical plan, and the end-to-end user journey.

Key decisions

Document AI boundaries, language/accessibility choices, data sources, and implementation trade-offs.

Outcome & next steps

Add prototype or rollout evidence, farmer feedback, delivery results, and the next product milestone.

Suggested evidence: a farmer journey map, key app screens, system flow, team responsibilities, and validation notes.

03 / CASE STUDY FRAMEWORK

Eloqui Android Tutor

An offline-first English vocabulary companion with multi-AI support.

RoleSolo developer
StackKotlin, Compose, Room, MVVM
FocusReliable personal learning

Context & challenge

Introduce the learning problem and why an offline-first experience matters.

Goal & success signals

State the intended learning behaviour and how a useful tutor experience is evaluated.

Approach

Describe the learning loop, Android architecture, data model, and AI integration plan.

Key decisions

Show the reasoning behind offline storage, MVVM, background work, and multi-AI support.

Outcome & next steps

Add build status, testing insights, learning outcomes, and the roadmap for the next release.

Suggested evidence: vocabulary flow, Compose screens, data/architecture diagram, test results, and a product reflection.