BLACK BOX / DANTE-1

I build production systems where product interfaces, backend constraints,
data retrieval, security, and AI-assisted workflows all meet.

Current orbit: Switzerland · Senior Software Engineer · Full-Stack & Backend
Primary sectors: aviation · consumer traffic · retail operations · AI products

I'm Matteo Dante — a senior software engineer based in Switzerland with 8+ years across aviation, high-traffic consumer platforms, CRM / sales analytics, retail operations, and solo-built App Store products.

I like systems that have to work outside the demo: restricted enterprise environments, real users, paid subscriptions, search quality problems, observability, security, and UI surfaces that people actually operate every day.

• At Pilatus Aircraft: one of the principal engineers on a greenfield pilot-training platform shipped on-premise into enterprise customer environments.
• Shipped GymTree: full-stack fitness app live on the App Store — coach web, trainee mobile app, background workers, subscriptions, and production AI for coach chat + personalized plans. Open App Store
• Shipped Maestro: native iOS AI tutor live on the App Store — source-grounded courses, lessons, quizzes, review, and tutor chat. Open App Store
• Open-source signal: claude-local-docs, a Claude Code plugin for local-first documentation retrieval with hybrid search and reranking.
• This profile's visual language: lifted from portfolio-cockpit, my 3D portfolio where you fly between planets instead of scrolling sections.

Greenfield pilot-training platform aggregating simulator, quiz, and lesson-system data. Enterprise on-premise delivery, regulated quality process, waterfall delivery model, microfrontend architecture, frontend design system ownership, reviews, hiring interviews, and onboarding.

Backend microservices for a high-traffic consumer directory platform with subscription advertisers, millions of pageviews per day, 2M+ registered users, and 10k+ peak concurrent traffic.

LLM agents, streaming responses, structured outputs, persistent AI chat, embeddings, BM25, Reciprocal Rank Fusion, and cross-encoder reranking — used for real product flows, not just demos.

Maestro is a native iOS AI tutor on the App Store. GymTree is a fitness app live on the App Store. Both are end-to-end product work: app UX, backend/API, subscriptions, observability, store presence, and production AI flows.

• Maestro: https://apps.apple.com/us/app/maestro-learn-anything/id6780267046?uo=4
• GymTree: https://apps.apple.com/us/app/gymtree-workout-ai-coach/id6761392403?uo=4

┌─ HOW I LIKE TO SHIP ───────────────────────────────────────────────┐
│ 01. Product UI should feel intentional, not generated.             │
│ 02. Backends should be observable, secure, and boring under load.   │
│ 03. AI belongs where it removes real workflow friction.             │
│ 04. Retrieval quality beats prompt theater.                         │
│ 05. A side project is not done until it survives production shape.  │
└────────────────────────────────────────────────────────────────────┘

• Portfolio / AI assistant: matteodante.it
• LinkedIn: linkedin.com/in/matteo-dante
• Email: matteo.dante659@gmail.com