I'm a Software Developer and Automation Engineer from Pakistan. I make useful software that helps businesses use modern AI tools better, do less manual work, and save time.

My recent work focuses on applied AI and intelligent automation, especially using open weight AI models in real projects. I enjoy taking vague or messy requirements and turning them into systems that are simple, reliable, and easy to maintain.

• Automation and RPA using Python, from small scripts to full workflows
• Web scraping and data collection from static and dynamic sites using Playwright and modern browser automation tooling
• Applied AI solutions (open weight models)
  • Product similarity, visual search, and image understanding with Qwen3 VL, Qwen3.5, and local vision models like Gemma 3 and Llama 3.2 Vision deployed via llama.cpp
  • Background removal and image editing with Qwen Image Edit
  • Image and thumbnail generation with Z Image Turbo, FLUX.1 dev, and FLUX.1 Kontext dev via ComfyUI
  • Qwen3 Coder and similar open weight models for agentic tooling and AI assisted automation
• Web and API development with FastAPI, WebSockets, and simple frontends in React or Vue when needed
• End-to-end testing and QA using Playwright and Appium for production ready automation
• Systems and graphics programming in C++ and Python, including Windows system-level tooling, real-time graphics pipelines, and GPU compute with HLSL and CUDA
• 3D tooling and Blender addon development: building Blender addons and Python scripts that automate digital twin production pipelines, covering mesh cleanup and optimization, quad conversion of AI generated geometry (from tools like Trellis and Hunyuan3D), material deduplication, pipe detection, mesh joining, and parametric structure generation from blueprint specifications

• Building client focused AI tools that solve real business problems
• Turning AI models into usable features, not just demos
• Exploring automation of digital twin creation workflows, including custom tooling and addons for 3D modeling software, using AI models to generate 3D geometry and estimate real-world dimensions from single images, and generating structures programmatically from blueprint specifications
• Open to freelance, contract, or long term roles involving AI applications, automation, or data driven systems

Almost everything I build is written by hand, line by line, with a lot of care and time put into every project. Over the past few months I have started integrating AI assistance tools like Claude to improve certain parts of older software or add new features on top of an existing foundation. Whenever that is the case, it will be clearly stated in the repository. I think being upfront about it is part of writing honest, maintainable software.

Website: https://danyalziakhan.com
GitHub: https://github.com/danyalziakhan
LinkedIn: https://www.linkedin.com/in/danyalziakhan
Dev.to: https://dev.to/danyalziakhan
Facebook: https://www.facebook.com/danyal.zia.khan

I like building software that feels useful the moment you start using it. If you're looking for someone who can turn ideas, data, or AI models into working systems, feel free to reach out.