Note: The following post was originally posted on Medium on Sept. 8, 2019 When we first started deploying neural network models in production at Talentpair, they were small (-ish), they only ran occasionally, and most importantly there were only a few of them. But as one proved its worth, another soon followed and our naive … Continue reading A Keras Meta Model Served
Creating and Using Custom Spatial Data with Django, Postgres, and GeoDjango
At Talentpair, we decided it would be helpful to track the commute preferences for our candidates. City dwellers often have opinions about which neighborhoods they're willing to commute to, and knowing these commute preferences allows us to make better matches for candidates and jobs. This post will cover how to: Create a minimal database tableCreate … Continue reading Creating and Using Custom Spatial Data with Django, Postgres, and GeoDjango
Serving Word Vectors for Distributed Computations
Word vectors are amazing! Vectors for everything! And that is why we use them for various deep learning tasks at Talentpair. What makes them amazing? Each word in the model is represented by vector. Each vector captures the semantic meaning of the word. Therefore, related terms will be close to each other, unrelated terms faraway. … Continue reading Serving Word Vectors for Distributed Computations
Visualizing Convolutional Nets for NLU Tasks
With deep learning becoming the foundation for Talentpair’s matching algorithm, we wanted to get a better understanding how our convolutional neural networks make the decisions they do. Google’s TensorBoard has been a tremendous help to achieve our goal (read more about Talentpair’s deep learning stack here). Unfortunately, due to our model setup, we couldn’t take … Continue reading Visualizing Convolutional Nets for NLU Tasks
Talentpair’s Deep Learning Stack
Over the past few months, we have upgraded our deep learning stack to take advantage of the latest developments in the world of deep learning. We transitioned from Theano to Tensorflow, upgraded our Keras version, introduced GPUs into the Talentpair stack and rolled out helpful machine learning tools to accelerate our model iterations. Transitioning from … Continue reading Talentpair’s Deep Learning Stack
Running AngularJS Tests with Jest
Earlier this year, we began the process of using React in our AngularJS app. Luckily, rendering React components within Angular isn’t too hard. We use a component “bridge” based on the approach shared by the Small Improvements team, which is working well. A side-effect of bringing React into the codebase was seeing how fast Jest … Continue reading Running AngularJS Tests with Jest
Django Multiprocessing
Once we got to a certain scale, it became necessary to start multithreading many of the scripts we run to keep our infrastructure healthy. Turns out that does not work very well out of the box with Django (we are currently running 1.11 on Python 2.7 though py3 is coming to our stack soon!). We … Continue reading Django Multiprocessing