I build software that scales, from distributed cloud backends ingesting 50k+ records per upload, to a farmer-facing platform serving 11,000+ users across low-connectivity regions of Africa.
With 8+ years across cloud, backend, mobile, and AI-integrated systems, I work at a rare intersection: I can architect the distributed backend, design the cloud infrastructure, integrate AI orchestration, and ship the client that consumes it. That full-stack depth means I move faster, mentor broader, and own more of the problem end-to-end.
Currently → Senior Software Engineer @ JustDiggit (NGO)
Focus → Cloud Infrastructure · Agentic AI Orchestration · Data Systems
Location → Nairobi, Kenya · Remote-first · Teams across UK, US, Kenya & Tanzania
Open to → Senior/Staff Engineer · Engineering Lead · Remote or Kenya-based
Languages & Runtimes
Backend & API
Messaging & Event-Driven
Cloud & Infrastructure
AI & Data
-00B4D8?style=flat-square&logoColor=white){kind=icon}
Mobile & Desktop (Cross-Platform)
AI-Assisted Development
Backend and cloud architecture for a Kotlin Android platform serving 11,000+ farmers across low-connectivity regions of Africa, with active users across Africa.
- Design and manage Azure infrastructure as code (Bicep), ensuring reproducible, version-controlled environments for field-constrained deployment conditions across multiple countries
- Own the data and analytics architecture across Firebase, Google Analytics, and BigQuery for reliable program and donor reporting
- Leading migration of a companion Angular CMS/admin portal to Blazor, consolidating the platform onto a single stack
C# .NET Azure Bicep MySQL BigQuery Blazor Firebase
Multi-tenant cloud billing platform for enterprise clients, integrating AWS and Azure with real-time visibility.
| What | Result |
|---|---|
| Manual invoicing automated via Node.js/TypeScript microservices | 60% reduction |
| Billing accuracy across enterprise clients | 95%+ |
| Fault-tolerant ingestion pipelines per upload | 50k+ records |
| Job failure rate (Redis caching + retry strategies) | ↓ 40% |
| Environment provisioning time via Terraform IaC | Days → minutes |
| MTTR via full-stack observability & alerting | < 30 minutes |
Node.js TypeScript C# PostgreSQL Redis AWS Azure Terraform Docker Kubernetes SignalR Azure Service Bus
Microsoft Teams chatbot using Azure OpenAI, LangChain, and vector databases to answer questions from a SharePoint document repository, deployed for global bid and proposal teams.
- Designed ETL pipelines for document ingestion and vector embedding, indexing SharePoint-hosted knowledge base content
- Enabled semantic search over large unstructured enterprise knowledge bases directly within Teams
- Reduced internal bid preparation time by 40%, improving decision-making speed for cross-functional teams
C# .NET TypeScript Azure OpenAI LangChain Pinecone Milvus Azure Functions
Real-time multilingual translation system spanning WPF, UWP, Xamarin.Forms, and .NET MAUI, supporting 43+ languages for enterprise clients in the UK and US.
- Built low-latency audio streaming using Windows.Media, AVFoundation, and Android.Media, achieving sub-100ms processing jitter
- Implemented SignalR-based real-time messaging for live translation updates across all platforms
- Migrated legacy Xamarin.Forms and native Kotlin Android codebases to .NET MAUI, unifying the team onto one shared codebase and cutting maintenance overhead by 50%+
C# .NET MAUI WPF UWP Xamarin.Forms SignalR Azure Cognitive Services Redis
High-performance club management platform serving 10,000+ users across mobile and web, backed by a scaled ASP.NET Core backend.
- Built offline-first SQLite sync engine with EF Core for resilient cross-platform data consistency
- Integrated SignalR for sub-second distributed state synchronization: reservations, live seat maps, updates
- Scaled ASP.NET Core backend for 100k+ monthly transactions, reducing query latency by 45% via SQL optimization
- Designed OIDC/OAuth2 authentication with RBAC for secure multi-role access control
C# .NET MAUI ASP.NET Core SignalR SQLite Entity Framework Core PostgreSQL Redis
| Metric | Result |
|---|---|
| Farmers served (Kijani, JustDiggit) | 11,000+ |
| Manual invoicing automated | 60% |
| Billing accuracy | 95%+ |
| MTTR after observability | < 30 min |
| Azure cloud spend saved | 25% |
| Bid preparation time cut (RAG) | 40% |
| Maintenance costs post-MAUI migration | 50%+ lower |
| Terraform IaC provisioning | Days → minutes |
| Unit test coverage improved | 20% → 85% |
| SQL query latency reduced (ClubFusion) | 45% |
| Certification | Issuer | Status |
|---|---|---|
| AWS Certified Developer – Associate | Amazon Web Services | ✅ Active |
| Microsoft Certified: Azure Developer Associate | Microsoft | ✅ Active |
| Microsoft Certified: DevOps Engineer Expert | Microsoft | ✅ Active |
| Microsoft Certified: Azure Fundamentals | Microsoft | ✅ Active |
| Docker Certified | KodeKloud | ✅ Active |
"The best systems are the ones nobody notices, because they just work."
I write code for the engineer who maintains it next, design for the user who never reads docs, and architect for the incident that hasn't happened yet. I use AI tools as a force multiplier, not to write code blindly, but to move faster on the parts that don't need my full attention, so I can bring more of it to the parts that do.
I'm always open to conversations about distributed systems, cloud architecture, agentic AI, or remote engineering roles, especially where technology serves real-world, resource-constrained environments.
Nairobi, Kenya · Open to remote roles globally · Building things that scale since 2018