Chun-Ho Hung and Nikhil Garg discuss Quanta, Quora's counting system powering their high-volume near-real-time analytics, describing the architecture, design goals, constraints, and choices made.
Mattia Battiston shares from his experience at Sky Network: what metrics they use, how they use them, what pitfalls they encountered and what little data they collect to get a whole lot of value.
Nikhil Garg talks about the various Machine Learning problems that are important for Quora to solve in order to keep the quality high at such a massive scale.
Bo Peng goes over how Datascope iterated on the major pieces of the Expert Finder application project to produce actionable insights and recommendations on methodologies.
Shakhina Pulatova overviews the Instant Search experience at LinkedIn and how they use Machine Learning to deliver personalized results as the query is typed.
Preslav Le talks about how Dropbox’s infrastructure evolved over the years, how it looks today, as well the challenges and lessons learned on the way.
John Rizzo introduces Twitch's chat's architecture, telling how their engineers investigated and worked through the issues in what turned out to be a make-or-break situation for the company.
Haley Tucker discusses how other systems may affect Netflix' services, strategies to protect their systems and make sure they won't fail even if things go wrong.
Matt Ranney talks about the limits that some companies have encountered in their large microservices deployments and some non-microservices approaches to those same problems.
Keith Adams tours Slack's infrastructure from clients into the Slack datacenter, and around the various services that provide real-time messaging, search, voice calls, and custom emoji.
Josh Evans talks about the chaotic and vibrant world of microservices at Netflix, exploring the cultural, architectural, and operational methods that lead to microservice mastery.
Neha Narkhede shares the experience at LinkedIn moving from ETL to real-time streams, the challenges of scaling Kafka to hundreds of billions of events/day, supporting thousands of engineers, etc.