techxiv
Slack Bug Bounty: Three Years Later 
Emergence of grounded compositional language in multi-agent populations - Prediction and control with temporal segment models
The end-to-end refresh of our server hardware fleet - Designing 100G optical connections
- The software ecosystem for storage hardware
- Third-person imitation learning
- Reducing Slack’s memory footprint
- Attacking machine learning with adversarial examples
- End-to-end optimizations for dynamic streaming
- Spatial audio — bringing realistic sound to 360 video
A glimpse into GitHub’s Bug Bounty workflow 
Jane Street Tech Talks: Verifying Puppet Configs - Machine Learning @Scale 2017 recap
- Adversarial attacks on neural network policies
- Search at Slack
- Using Apache Spark for large-scale language model training
- Making Slack Faster By Being Lazy: Part 2
- Building scalable systems to understand content
- Disaggregate: Networking recap