25+ years solving complex problems across software, infrastructure, and AI — whatever stack the problem lives in. I don't specialize in a language. I specialize in finding answers.
Currently focused on local LLM deployment, AI integration architecture, and self-hosted AI systems built on consumer hardware.
Django 5 Docker orchestration platform with REST API, WebUI, and LLM-assisted log triage, GPU reporting, and service mapping. Published to GitHub Container Registry with automated CI/CD.
Django • FastAPI • Docker • PostgreSQL • Ollama
View RepositoryProfessional-grade application for managing AI model training across multiple GPU workers. Supports QLoRA, Unsloth, RAG, and standard fine-tuning with real-time epoch and loss monitoring.
FastAPI • React • TypeScript • SQLAlchemy • Ollama
View RepositoryFull-stack Laravel platform implementing a RAG pipeline for resume analysis, skill extraction, and AI-driven job matching. Features PDF/DOCX parsing, Redis queues, admin panel, Docker orchestration, and 110 tests / 370 assertions.
Laravel • PHP • Ollama • RAG • Docker • Redis • MySQL
View RepositoryFull-stack Python/React application that uses a local Ollama LLM to simplify complex topics into plain language — designed for users with limited domain knowledge.
Python • Flask • React • Ollama
View RepositoryFlask-based chat interface backed by locally-running Ollama models. JSON-configurable model switching, typed client wrapper, and full pytest coverage.
Python • Flask • Ollama • pytest
View RepositoryEngineered cross-database synchronization reducing MS SQL ↔ MySQL sync time from ~1.5 hours to ~5 minutes across a mixed Ubuntu + Windows Server environment.
MS SQL • MySQL • Linux • Windows Server
Refactored legacy ERP workflows into Laravel, migrated MyISAM to InnoDB, and built 10+ reporting tools — contributing to ~40% operational efficiency improvement.
Laravel • PHP • MySQL • MS SQL
Available for remote consulting engagements and employment. If you have a hard problem that needs solving, that's what I do.