Turaboy H.

I'm a senior full-stack engineer with 10+ years of commercial work, and for the last few years most of it has involved putting AI and ML into production systems that real businesses depend on. Some of that work: an ML pricing engine for a Japanese airline that forecasts demand across thousands of flights a day. A housing analytics platform for Weiss Residential Research with AI-scored property imagery and search across every US ZIP code. A geospatial planning tool for King County covering 90,000+ roads. Fortescue Metals' enterprise project portfolio system, where I led the migration to Vue 3 and .NET 9. Marketing analytics for Kraft Heinz. Four years modernising applications for the Pennsylvania Department of Transportation. 1,300+ hours on Upwork, 100% Job Success, Top Rated Plus. WHAT I DO - Full-stack delivery: Python (FastAPI, Django) or C#/.NET backends, Vue 3 or React frontends, PostgreSQL, Docker, CI/CD - ML in production: demand forecasting, pricing, valuation models (XGBoost, scikit-learn), scheduled retraining on Kubernetes - LLM integration: Claude and OpenAI APIs, RAG with pgvector, tool use, evals, guardrails - Data and geospatial: PostGIS, MapLibre GL, Elasticsearch, Snowflake, SSIS pipelines - Legacy modernisation: ASP.NET Web Forms to React/.NET Core, Vue 2 to Vue 3, without breaking production RECENT WORK - Solaseed Air (Japan): XGBoost models for bookings, load factor and ADR; pricing recommendations built on competitor and event signals; daily retraining on Kubernetes - Talent International / Enco Group (Perth): lead developer on Fortescue's Catalyst PPM platform; led the Vue 3 + .NET 9 migration, Snowflake data pipelines, earned-value and purchase-order tracking - King County (Seattle): stormwater retrofit planner on Next.js + FastAPI with MapLibre; 60% faster data loads via Web Workers and LocalForage caching; retained for follow-on work - Weiss Residential Research (Boston): Django REST API over nationwide housing data, Elasticsearch search, embeddable React widgets, AI-scored image pipeline, two-year engagement - Kraft Heinz (Chicago): Kraft-O-Matic marketing analytics platform, Django + Vue, Azure AD SSO, Shopify integration - PennDOT (Harrisburg): four applications rebuilt from Web Forms to React + .NET Core across multiple bureaus WHY CLIENTS COME BACK - You get a working feature with UI, tests and deployment, not a prototype - I've worked inside mining, government, healthcare and Fortune 500 environments, so I'm used to compliance, reviews and strict requirements - My workflow runs on Claude Code with review, testing and security subagents, so I move fast without cutting corners STACK: Python (FastAPI, Django) · C#/.NET · Vue 3, React, Next.js, TypeScript · PostgreSQL, PostGIS, pgvector, SQL Server, Snowflake · Redis · Docker, Kubernetes · AWS, Azure, Azure DevOps · Claude API, OpenAI API · XGBoost, scikit-learn Send me what you're building, or what's broken. You'll get a concrete approach, usually within a few hours. Currently based in Perth, Australia (AWST), working with clients across US, UK and Australian time zones.