What Is the VAF UBWENGE TECH Intelligence Lab?

More Than a Side Project

When we started VAF UBWENGE TECH, we knew that Easy GO was one application of a much larger set of questions we wanted to explore. How do you build AI systems that work reliably in low-resource environments? How do you collect meaningful training data in markets where data infrastructure is nascent?

Those questions needed a dedicated space. That space is the Intelligence Lab.

What the Lab Actually Does

The Intelligence Lab is the research and development function of VAF UBWENGE TECH. It is where we prototype ideas before they become product features, run experiments that might fail, and document what we learn whether or not it leads to something we ship.

Current active research areas include geospatial AI for urban mobility in African cities, lightweight ML models optimized for inference on low-spec mobile devices, and natural language interfaces in Kinyarwanda for non-English-speaking users.

The Connection to Easy GO

Everything the Intelligence Lab researches eventually finds its way into Easy GO or into tools we share publicly. The price prediction engine powering Easy GO today was first prototyped in the lab. The driver demand forecasting model we are currently testing will become a product feature when it meets our accuracy threshold.

This feedback loop between product and research is intentional. Research without application drifts. Product without research stagnates. The lab keeps our product honest and our product keeps our research grounded.

Why Rwanda, Why Now

Rwanda has made deliberate investments in becoming a technology hub for East and Central Africa. The infrastructure — reliable power, growing 4G coverage, a young and educated population, and a government that actively supports innovation — creates conditions where serious technical work is possible.

We are building the Intelligence Lab here because this is where the problems we care about are most acute and where the solutions we develop will have the most direct impact.

What We Plan to Publish

We are committed to open research. As our work matures, we plan to publish datasets, model architectures, and findings that other builders in the region can use. The problems of African urban mobility are too large for any one team to solve.

If you are a researcher, engineer, or student who wants to collaborate with the Intelligence Lab, we want to hear from you. The door is open.