Skip to content
View 2troll's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report 2troll

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
2troll/README.md

Tony Hanma

Software engineer — internationalization, RTL interfaces, and localization done right.

I build multilingual products end to end: the engineering and the language layer, so nothing gets lost between them.


What I do differently

Most localization breaks in the same place — text machine-translated and dropped into a layout that was never built for it. I build interfaces that are internationalized from the ground up: mirrored right-to-left layouts, bidirectional text that doesn't scramble, correct numeral and date handling, and typography that holds up — then I handle the language layer myself instead of outsourcing it.

Not a developer who bolts translation on at the end, and not a translator who can't touch the code.


Selected work

Project What it is
Kuruma OS Dealership & vehicle-export management desktop app (Windows/macOS). Large vehicle catalogue, sales & CRM, export documentation, invoicing, OBD-II diagnostics, multi-office sync. Offline-first, bilingual UI.
Kansai Radar Autonomous daily event bot. Runs free on GitHub Actions, collects from keyless public sources, geocodes with OpenStreetMap, and pushes to Telegram. No servers, no database, no paid APIs, no LLM.
trackVideo Give it a driving video and it works out where it was filmed — reading plates, signs and postcodes with OCR, then mapping the evidence. Fully local: yt-dlp + ffmpeg + Tesseract + OSM.
akiya-finder Property-listing finder. Scrapes public listings into local SQLite, geocodes addresses, and plots them on a Leaflet map with configurable filters.
Kei-Mecanico PWA that uses the phone camera to identify vehicle parts and paperwork and explain them on the spot. Node backend, Docker, deployed on Render.
TOEIC Simulator TOEIC Listening & Reading exam simulator. React + Vite, authentic item types, strict exam timing, and three-tier feedback on every question.

Stack

Python, JavaScript, TypeScript, Node, React, Java, C# SQLite, Docker, GitHub Actions, Tailwind, PWA, Capacitor OpenStreetMap / Leaflet, Tesseract OCR, i18n and RTL

A recurring constraint in my work: free and self-hosted by default — no paid APIs, no vendor lock-in, works offline where it can.


Background

Practical Engineer in Software Engineering, graduated with distinction.


Open to work

Remote. Contract or freelance, project-based or retainer. Particularly interested in multilingual products, localization, and travel tech.

Reach me by opening an issue on any of my repositories.

Pinned Loading

  1. akiya-finder akiya-finder Public

    Property-listing finder. Scrapes public and municipal listings into local SQLite, geocodes addresses, and plots them on a Leaflet map with configurable filters.

    HTML

  2. kansai-radar kansai-radar Public

    Autonomous daily event bot. Runs free on GitHub Actions, collects from keyless public sources, geocodes with OpenStreetMap, and pushes to Telegram. No servers, no database, no paid APIs, no LLM.

    Python

  3. kei-mecanico kei-mecanico Public

    Installable PWA that uses the phone camera to identify vehicle parts and paperwork and explain them on the spot. Node backend, Docker, deployed on Render.

    JavaScript

  4. kuruma-os-releases kuruma-os-releases Public

    Dealership & vehicle-export management desktop app (Windows/macOS). Vehicle catalogue, sales & CRM, export documentation, invoicing, OBD-II diagnostics, multi-office sync. Offline-first.

  5. TOEIC-EXAM TOEIC-EXAM Public

    TOEIC® Listening & Reading exam simulator built with React + Vite. Authentic item types, strict exam timing and a mandatory three-tier feedback system on every question.

    JavaScript

  6. trackVideo trackVideo Public

    Give it a driving video and it works out where it was filmed — reading plates, signs and postcodes with OCR, then mapping the evidence. Fully local: yt-dlp + ffmpeg + Tesseract + OSM.

    Python