I'm a full-stack engineer with 8+ years in production, now focused on AI-native development. I spent five years at Pesapal, a multi-tenant payments SaaS, owning products end-to-end: backend services, REST APIs, dashboards, and CI/CD on Google Cloud Platform. Today I build production AI systems, shipping React/TypeScript frontends against Python and Node backends, with LLM agents and n8n pipelines that replace hours of manual work. I also built the RAG layer, per-tenant data isolation, and the self-hosted inference stack behind it, and I own all of it in production.
AI-assisted development is how I work, not something I've trialled. Claude Code, Cursor, and Copilot run through my whole build cycle, while I keep the judgment on architecture, failure modes, and what reaches production.
I'm comfortable inheriting a codebase, forming a view on it, and shipping fast without breaking things. I work independently and think in terms of product and systems.
Portfolio: https://captano.dev/
Sole engineer on a production browser voice agent (WebRTC, VAD, STT, LLM, TTS) achieving time-to-first-audio as low as 14ms on common turns. Built multi-tenant RAG retrieval with pgvector and row-level security. Self-hosted GPU inference (vLLM), cutting container cold start ~30%. Delivered outbound automation engine, React/TypeScript client dashboard, and billing service.
Led development across three payment products serving 30,000+ merchants in East Africa. Owned backend services, REST APIs, mobile apps, and dashboards. Optimized queries to reduce latency and database load. Implemented CI/CD on Google Cloud Platform, reducing deployment time. Worked across Kotlin/Jetpack Compose (Android) and Swift/SwiftUI (iOS) alongside Python and Node backends.
This professional hasn’t added portfolio projects yet.
This professional hasn’t listed any services yet.
