A software developer who loves building *things for software developers.


Built with Python and Django, running on Gunicorn. Uses Wordnet and some NLTK modules for semantic categorisation.
Built on top of Tornado and React. Trained data stored on a Redis database with a prefix. Uses web socket for real-time monitor on the client-side, and Pub/Sub interface of Redis with a Tornado endpoint.
Built with React and Redux. The state is persisting on the browser's local storage with a Redux middleware. Language interpreters are running on web workers. Workers are spawming on demand. Used gulp on development mode.
Built with Django and Backbone.js, and socket.io for realtime updates. Used MongoDB for storing newsfeed, notifications and documents. The rest of the data storing on relational database.
It was a weekend project. I built it with Backbone.js. Supports step by step debugging, delay time between instructions, shows pointer position with caret by instantaneously.
Built with Backbone.js. It allows extend the toolset with an internal declarative API. The data layer is loosely coupled, It's easy to integrate with a RESTful API.
The graph is rendered on Gephi app. I used D3.js for drawing nodes on the browser. I feed the gml file with a python script that uses NetworkX for building graph structure.
Dictdiffer is a helper module that helps you to diff and patch dictionaries in Python. The project is maintaining and developing by Invenio Software now.
I used Bottle and Pymongo libraries. It can be used as a temporary backend service for your persistance needed apps. Supports module system and have some builtin modules.
Parses nginx configuration files. The parser written with Pyparser which is a python library that allows you to generate parsers with own internal DSL.
It uses Python's AST (abstract syntax trees) module. Parsers returns the skeleton of provided django models. It's using on dbpatterns for generating diagrams from django models.