Understand
Clarify the decision, user, source systems, constraints, and definition of a useful result.
Professional profile
The dashboard is the easy part.
Getting the numbers to a place where they can be trusted takes longer, and it's the part most projects underestimate. That's where most of my work goes.
At a glance
- Solutions Architect
- Kampala, Uganda
- React · Python · SQL
- IBM Data Science
Who I am
I work across the whole stack, and I like it that way.
I'm Okurut Maurice Leonard, a solutions architect and full-stack engineer in Kampala, Uganda. I take on the whole delivery: architecture, React interfaces, Python and Node services, SQL systems, reporting, machine learning, and automation. Handing a project between four suppliers is where most of the detail gets lost.
At Logos Cloud Services I build pipelines, reporting workflows, and model-ready datasets for SME clients in fintech, agribusiness, and logistics. I also collaborated on Aristoc Booklex, a live e-commerce platform for an established Ugandan bookstore chain, where I worked across the full stack.
Earlier local-government work taught the rest: handling official records carefully, checking integrity before reporting, and explaining findings in plain language.
What guides the work
- Reliability before presentation. A polished dashboard cannot compensate for incomplete or inconsistent source data.
- Context before tooling. The reporting question and operating process determine the right technical approach.
- Clarity for the end user. Documentation and data storytelling are part of the solution, not an afterthought.
- Maintainability after delivery. Repeatable workflows, checks, and handover notes make the work usable beyond a single analysis.
Working approach
The same four steps, whatever the problem is.
The technology changes constantly. This sequence hasn't needed to.
Structure
Map fields, define consistent formats, and establish validation rules before analysis.
Build
Implement the pipeline, database, report, dashboard, or automation in testable steps.
Handover
Document assumptions, checks, and recurring tasks so the solution remains understandable.
Relevant toolkit
The toolkit, and where each part earns its place.
Everything listed here is backed by either paid work or a repository you can open.
Web & interfaces
JavaScript and TypeScript, React and Next.js, semantic HTML and CSS, responsive and accessible interface work.
Services & APIs
Python with Django, Flask, and FastAPI, Node services, REST design, authentication, and third-party integration.
Data & storage
SQL databases and modelling, pandas and NumPy, pipelines, validation, Power BI, and Python visualisation.
Applied modelling
Scikit-learn, TensorFlow, Keras, data preparation, classification, regression, and evaluation.
Delivery
Git and version control, environment configuration, deployment, documentation, and handover notes.
Architecture
System and data modelling, API contracts, integration boundaries, and written decision records.
Need one person who can cover the whole build?
Read through the capabilities, or just tell me what you're trying to solve.