« Smerity.com
Articles
2018
October 18, 2018
The compute and data moats are dead
August 4, 2018
What to pay attention to in the OpenAI Five Benchmark
April 6, 2018
T-Mobile Educational Material on Plaintext Passwords and Computer Security
February 25, 2018
Cython - making Python high and low level
January 8, 2018
The Turing test isn't won by machines, it's lost by humans
2017
December 11, 2017
Bias is not just in our datasets, it's in our conferences and community
November 19, 2017
Understanding the Mixture of Softmaxes (MoS)
November 3, 2017
Backing off towards simplicity - why baselines need more love
February 26, 2017
Stop saying DeepCoder steals code from StackOverflow
February 23, 2017
Cloudflare's Cloudbleed in plain English
2016
November 17, 2016
Peeking into the neural network architecture used for Google's Neural Machine Translation
November 5, 2016
Hunting through the ICLR 2017 submissions
July 30, 2016
It's ML, not magic: machine learning can be prejudiced
July 3, 2016
It's ML, not magic: simple questions you should ask to help reduce AI hype
June 27, 2016
Explaining and illustrating orthogonal initialization for recurrent neural networks
June 11, 2016
In deep learning, architecture engineering is the new feature engineering
March 25, 2016
TayAndYou - toxic before human contact
2015
October 30, 2015
Why small files are a curse for big datasets
September 8, 2015
The limitations of the magic undelete in Linux
August 5, 2015
Question answering on the Facebook bAbi dataset using recurrent neural networks and 175 lines of Python + Keras
May 25, 2015
How Google Sparsehash achieves two bits of overhead per entry using sparsetable
April 3, 2015
The lack of HTTPS at Amazon: identifying items purchased using information leakage
2014
September 24, 2014
The Saga Of: Shuffling without using memory
May 23, 2014
Of hacker schools and degrees
March 20, 2014
Navigating the WARC file format
2013
December 26, 2013
Google, make Google Analytics HTTPS by default
October 6, 2013
Burning bits in the Digital Libraries of Alexandria
September 2, 2013
Co-founder Half-life
June 22, 2013
Dithering: how it works and why it matters
May 26, 2013
Where did all the HTTP referrers go?
May 12, 2013
The Wrong Type of Karma
April 20, 2013
Starting The Matasano Crypto Challenges
April 14, 2013
Big Ints Between Python, PyPy, Go and Java
March 21, 2013
The Mystery of the Spotty Animated GIF
2012
August 9, 2012
Facebook, for the sake of speed, fake it like Instagram!
August 8, 2012
Installing Vanilla Forums with performance in mind
June 7, 2012
Password Security: Why salting with usernames is no good
May 15, 2012
No magic: why everyone should program
April 27, 2012
The Dining Cryptographer's Problem
April 24, 2012
Fast Facebook OAuth Login
April 23, 2012
The Facebook (Maybe) API
April 20, 2012
A Lean Life
April 12, 2012
A Light Benchmark of Hetzner's 512MB VPS
April 3, 2012
Eric Ries & The Lean Startup Methodology
February 19, 2012
PJAX and Flask
2011
December 15, 2011
Chromium Fractal plated Heisenbugs, Mandelbugs and Schrödinbugs
December 2, 2011
Shift-Reduce Parsing with a Graph-Structured Stack
November 29, 2011
404 Marketing backed by Programming Zen
July 18, 2011
The First Honours Hurdle
July 11, 2011
Preliminary Timeline for Honours
2010
February 1, 2010
Google Internship Reflection
2009
October 26, 2009
Programming Interview Questions: mind teasers in binary
2008
June 1, 2008
Python to Java - A Survivalist's Guide
Popular Articles
Backing off towards simplicity - why baselines need more love
Peeking into the neural network architecture used for Google's Neural Machine Translation
It's ML, not magic: machine learning can be prejudiced
It's ML, not magic: the rise of AI-prefix investing
In deep learning, architecture engineering is the new feature engineering
How Google Sparsehash achieves two bits of overhead per entry using sparsetable
Where did all the HTTP referrers go?
Interested in saying hi? ^_^
Follow @Smerity