About

I'm Vincent Ndegwa, a software and AI engineer. For six years I have built production software used by teams in retail, finance and the creator economy, and today I focus on scalable applications and the AI features that make them more useful.My backend work spans PHP (Laravel), Node.js (NestJS and Express) and Python (FastAPI and Django). On the frontend I build with Vue (Nuxt) and React (Next.js). Much of my work is turning slow, tangled systems into fast, dependable services: tuning databases, introducing queues and caches, and shipping APIs that hold up at scale.At Paysoko Systems I lead engineering on a multi-tenant ERP and POS platform, from high-volume MySQL reporting to Stripe and Stripe Connect flows for split payments and subscriptions.

The AI thread

I studied for an MSc in Artificial Intelligence at the Open University of Kenya. My AI work covers NLP pipelines, LLM orchestration, retrieval-augmented generation and predictive analytics, always applied to real business workflows.It is practical engineering rather than experiments for their own sake: sentiment engines, recommendation systems and AI-driven automation that cut manual effort and speed up operations.

How I build

Clean architecture first. Explicit service boundaries, testable modules and observability from day one. A system that performs well but cannot evolve safely is a long-term risk.I take delivery seriously too. The systems I ship come with Docker environments, CI/CD pipelines and release practices built for distributed teams.

Education

• MSc Artificial Intelligence, Open University of Kenya (2023 - 2025)
• BSc Software Engineering, Kirinyaga University

Beyond the terminal

When I'm not in a codebase I'm running local LLMs with Ollama, following football, or writing notes to make sense of what I've learned.Browse my projects or book a call. I'm always open to interesting problems.