Machine Learning Engineer and Researcher. I like good old fashioned data structures and algorithms, algorithms for approximating hard counting problems or sampling from intractable probability distributions, C++, Lua, Python, Bash, Clojure, Emacs, functional programming, software architecture, representation learning, Bayesian machine learning, statistical language modelling and extracting information from unstructured data, e.g. text. At the moment, I am particularly interested in automated knowledge base construction, building knowledge bases for inference, weak supervision, the application of Deep Learning to Natural Language Understanding, and writing Lua/C/CUDA code with Torch7. I have been slowly converted to Agile development, and am now an enthusiastic supporter of using it in both engineering and research projects.