Senior Software Engineer | Digital Health | AI & Scalable Systems
I design and build production-grade software systems across healthcare, AI, and enterprise platforms. My work focuses on scalable architectures, data-intensive applications, and secure system design, particularly in digital health and EMR systems.
π« Open to senior engineering roles, technical consulting, and collaboration β reach me on LinkedIn or open an issue on any repo.
- π Building AI-powered health systems including diagnostic and risk assessment tools
- π§ Applying Machine Learning & Computer Vision to disease detection (mpox, imaging workflows)
- π₯ Developing EMR / HMS platforms with offline-first architecture and NDPR-compliant data security
- βοΈ Designing distributed systems, APIs, and cloud-native infrastructure
- Fullstack Engineering: Frontend, backend, and system design
- Cloud & DevOps: CI/CD, containerisation, infrastructure automation
- Digital Health Expertise: EMR systems, interoperability, health data pipelines
- AI Systems: Model training, inference pipelines, and real-world deployment
- Performance & Scalability: Optimised systems for high-load environments
Languages
TypeScript JavaScript Python Go PHP SQL Bash
Frontend
React Svelte SvelteKit Vue Angular Redux
TailwindCSS Bootstrap Sass HTML5 CSS3
HTMX Alpine.js
Backend & APIs
Node.js Express NestJS
Django Flask FastAPI
Laravel CodeIgniter
REST GraphQL
Databases & Caching
PostgreSQL MySQL MariaDB MongoDB
Redis SQLite
Cloud, DevOps & Infra
Docker AWS GCP Nginx Linux Jenkins
Messaging & Systems
Kafka RabbitMQ
Mobile & Cross-Platform
Cordova NativeScript Electron
AI / Data
ONNX Runtime InsightFace OpenCV CLIP
Machine Learning, Computer Vision, Inference Pipelines
Tools & Platforms
Git Postman Figma Firebase Heroku Appwrite
Bun PM2 Netlify
Fully offline, self-hosted media gallery with state-of-the-art face recognition and clustering β Google Photosβstyle people organization with no cloud, no telemetry, and no data leaving your machine.
- Local inference via ONNX Runtime: InsightFace
buffalo_l(SCRFD detector + ArcFace ResNet-50, 512-d embeddings) with CPU/GPU auto-selection - Multi-algorithm face clustering β agglomerative, DBSCAN, Chinese Whispers, HDBSCAN β plus multi-exemplar identity matching and intra-video face merge
- CLIP visual embeddings for similarity search and automatic tag suggestions
- FastAPI + HTMX/Alpine.js gallery over SQLite, with a WebP thumbnail cache and FFmpeg video keyframe sampling
github.com/RussellJapheth/kimeraΒ· AGPL-3.0 Β· 87 tests passing
A modern, offline-first spaced repetition flashcards app for learning languages, built with Svelte 5.
- SM-2-inspired scheduler with per-deck review queues and due-date spread (
src/lib/utils/srs.ts) - Offline-first: study data persists in the browser (IndexedDB) with a service worker and debounced background sync to a small JSON API
- Chinese learning focus: HSK-aligned decks, pinyin, tones, tone-sandhi detection, and cloze (sentence-fill) practice with multiple accepted answers
- Speech practice via the Web Speech API with accuracy evaluation, plus arcade games (Match Blitz, Number Rush) with combo scoring
- Progress system: XP, levels, streaks (with streak freezes), milestones, and weekly/monthly leaderboards
- SvelteKit + TypeScript + Tailwind CSS v4, deployed on Netlify Β· AGPL-3.0
A self-hosted, cross-device text and file sharing app inspired by Microsoft Edge Drop β single-user, minimal, and secure.
- Real-time sync over native WebSockets with a chat-style timeline, sticky day grouping, and lazy loading
- Text/link sharing, drag-and-drop multi-file uploads, clipboard paste, and EXIF-aware thumbnails via
sharp - Password auth with Argon2id + HTTP-only cookies; SQLite through Drizzle ORM (local or Turso) and local/S3-compatible storage
- SvelteKit + Tailwind CSS, PWA-installable Β· MIT
A simple, elegant Pomodoro-style work timer and todo list.
- Customizable interval timer paired with a persistent task list
- Fully client-side β tasks and settings stay in
localStorage, no account required - SvelteKit + Tailwind CSS + Bun, deployable to Netlify or any SvelteKit adapter Β· GLWTPL
AI-powered web platform for mpox risk assessment using image classification.
- Computer vision model for symptom detection
- Real-time inference pipeline
- Healthcare-focused risk scoring system
- Designed for scalability and field usability
A few other public repos:
- Rcopy β curl-based remote file copy with a CLI, real-time progress, and resumable downloads
- CheckIN β a streak calendar built with SvelteKit
- my-minimal-finance β privacy-focused personal finance tracker, fully in-browser (IndexedDB)
- codeigniter4-db-explorer β interactive terminal database explorer/editor for CodeIgniter 4
- Software Engineering
- Full Stack Development
- System Architecture
- Cloud Computing
- DevOps Engineering
- Machine Learning
- Computer Vision
- Digital Health
- Electronic Medical Records (EMR)
- Health Information Systems (HIS)
- API Design & Integration
- Distributed Systems
- Data Engineering
- Secure Systems (NDPR Compliance)
- π¬ Open to discussions on software engineering, AI, and system design
- π§© Strong focus on real-world problem solving and production systems
- π οΈ Building open-source, offline-first AI tools (see Kimera)
- βοΈ Preference: spaces over tabs


