Available for freelance & contract work

Full-stack engineer building production software from idea to deployment.

I design and build web applications, REST APIs, AI-powered tools, and cloud-based systems. My work spans React frontends, .NET and Python backends, PostgreSQL databases, and production infrastructure on AWS — from the first commit to the live URL.

Selected Work

Production applications built end-to-end — from architecture through deployment.

YASP

Open Source

Open-source API documentation and governance platform

YASP API Collections screen, showing an imported Swagger Petstore specification with endpoint count and last-updated timestamp

Problem

Developers need a fast, offline-capable way to manage, explore, and validate OpenAPI specifications without routing document data through a cloud service.

What I built

A full-stack API governance platform with an interactive spec editor (CodeMirror 6), a live Try It Out testing interface that proxies requests server-side to avoid CORS issues, AI-powered spec generation via OpenRouter, and Spectral linting running inside a Web Worker for non-blocking quality scoring. All spec data is stored in IndexedDB — no backend required for core functionality. Also ships as a desktop app via Tauri.

Key engineering details

  • Offline-first: all specifications stored locally in IndexedDB — no account required
  • Web Worker-isolated Spectral linting keeps the editor responsive under large specs
  • SSRF prevention on the API testing proxy with URL allowlisting and private-network blocking
  • AI spec generation and one-click fix suggestions via OpenRouter
  • Cross-platform desktop app (Tauri) alongside the web build, sharing the same core package

Stack

React 19 TypeScript React Router v7 Tailwind CSS v4 CodeMirror 6 Vite / Bun Spectral Tauri Vitest Playwright

WinService.Deploy

Open Source

Safe, scriptable deployment tool for Windows Services

Problem

Manual Windows Service deployments are error-prone: a failed copy or a service that won't restart can leave production in a broken, undocumented state with no clean way back.

What I built

A project-agnostic .NET 8 console tool that automates the full deployment lifecycle — stop, backup, copy, verify, restart — with automatic rollback if any step fails. Ships as a self-contained executable with an interactive Spectre.Console UI, a WhatIf dry-run mode, and full deployment logging for audit and troubleshooting.

Key engineering details

  • BLAKE3 file-hash verification after copy to catch silent transfer corruption
  • Automatic rollback to the last versioned backup on any failed step
  • Configurable retry logic for service stop/start with exponential backoff
  • WhatIf mode simulates a full deployment with zero filesystem or service changes
  • Retention-based backup cleanup to bound disk usage on the deploy target

Stack

C# / .NET 8 Spectre.Console Dependency Injection BLAKE3 Windows Service API

Reddit Research Tool

Production

RAG-powered intelligence tool for mining Reddit communities

Problem

Researchers and founders want to extract insight from subreddits — recurring pain points, emerging trends, common questions — without reading thousands of posts manually.

What I built

A Blazor Server application that crawls any subreddit via the Reddit API, chunks and embeds posts into a PostgreSQL vector store using Ollama, and exposes a real-time chat interface where users get cited, context-aware answers via RAG. Implemented real-time ingestion pipeline status over SignalR, per-IP rate limiting, lead capture with email notifications via Resend, and configurable retrieval parameters.

Key engineering details

  • Async streaming ingestion pipeline with live progress updates pushed via SignalR
  • Top-K vector retrieval reranked before LLM synthesis for higher answer quality
  • OpenRouter for configurable model selection and adjustable generation parameters
  • Per-IP rate limiting to prevent API abuse on the chat endpoint
  • Lead capture with email notifications (Resend) and PostgreSQL persistence

Stack

C# / Blazor Server .NET PostgreSQL pgvector Ollama OpenRouter SignalR Resend Docker Nginx

PDF Read Aloud

Live

Privacy-first PDF text-to-speech — documents never leave your device

PDF Read Aloud interface, showing the PDF upload panel and a synced reading pane for extracted text

Problem

Cloud TTS services require uploading document text to third-party servers — unacceptable for legal, medical, or confidential files. Privacy-preserving alternatives existed but required technical setup non-technical users couldn't navigate.

What I built

A local-first PDF reader that extracts text in the browser using PDF.js, streams it to an on-device Kokoro ONNX TTS model served by a FastAPI backend, and plays audio with sentence-level highlighting synchronized to playback. Built an adaptive prefetch engine that queues audio ahead based on current playback pace, a 150 MB IndexedDB audio cache that persists across sessions, and packaged the entire stack in a single Docker container.

Key engineering details

  • Zero document data sent to any external service — PDF processing is fully local
  • Adaptive prefetch engine: monitors playback pace and queues audio to eliminate stalls
  • 150 MB IndexedDB audio cache with LRU eviction for repeat listening without re-synthesis
  • Supports browser TTS (no setup) or local Kokoro ONNX (higher quality)
  • Sub-second playback start on cached audio; single-command Docker deployment

Stack

Python FastAPI Kokoro ONNX JavaScript PDF.js IndexedDB Docker Nginx

Technical Capabilities

Technologies verified from production projects.

Frontend

React 19 TypeScript React Router v7 Tailwind CSS v4 Radix UI Zustand Vite CodeMirror 6 JavaScript

Backend

C# / ASP.NET Core Blazor Server Python / FastAPI REST APIs Server-Sent Events SignalR Cookie Auth

AI & Data

Ollama (local LLM) pgvector RAG pipelines OpenRouter Kokoro ONNX TTS Semantic search Embedding models

Infrastructure

AWS Lightsail PostgreSQL Docker Nginx PM2 Linux Let's Encrypt

Integrations

Resend (email) Reddit API OpenRouter OpenAPI / Spectral Tauri (desktop)

Testing & Observability

Vitest Playwright React Testing Library MSW Serilog OpenTelemetry Prometheus

About

I'm a full-stack software engineer with a track record of taking software problems from initial requirements through architecture, implementation, and production deployment. I work across the full stack — React frontends, C# and Python backends, PostgreSQL databases, and Linux infrastructure on AWS.

My recent work has a strong AI and data component: building RAG pipelines with local embedding models and vector databases, integrating LLMs for document analysis and generation, and shipping privacy-preserving alternatives to cloud-dependent tools. I apply OWASP security practices throughout — from input validation and authentication to server hardening and security headers.

I'm available for freelance, contract, and consulting engagements. I can join a team for a specific build, take a product from zero to launch, or improve an existing system. I prefer projects where engineering quality matters and where there is a real problem to solve.

Have a project that needs to be built, improved, or shipped?

I'm currently available for freelance and contract work. Send me a brief description of what you're building and I'll get back to you promptly.