MLOps The Four Innovation Phases of Netflix’s Trillions Scale Real-time Data Infrastructure https://thedataexchange.media/building-machine-learning-infrastructure-at-netflix-and-beyond/ Real-time machine learning: challenges and solutions Why are ML Compilers so Hard? Compiler UCB CS294-113: Virtual Machines and Managed Runtimes Rendering I’ve been wanting to go back to my root of programming: games and rendering. A Review of Shader Language Finding Your Home in Game Graphics Programming » Read More

From HN

gRPC defined 18 status codes for returning different types of errors.

I think they’re a pretty good reference if you want to design clear error handling for API.

Context: Machine Learning Projects are mainly just a complex interdependent pipeline. We desperately need a better abstraction for them.

I’ve been taking naps multiple times a day. And surprisingly, a simple 5 minutes nap can help me avoid quite a lot of depression and anxiety.

Using AntiPatterns to avoid MLOps Mistakes, Nikil Muralidhar et. al.

I learned about this survey paper from The Data Exchange podcast. It’s a good introduction for people who just started deploying their machine learning project to production.

The paper focused on supervised learning, and forcasting applications. But the observation and recommendations should be general enough to be applied to other common machine learning deployments.

genuine teamwork in most organizations remains as elusive as it has ever been. organizations fail to achieve teamwork because they unknowingly fall prey to five natural but dangerous pitfalls The Natural pitfalls absence of trust: unwillingness to be vulnerable within the group. fear of conflict: incapable of engaging in unfiltered and passionate debate of ideas. They resort to veiled discussions and guarded comments lack of commitment: team members rarely, if ever, buy in and commit to decisions, though they may feign aggrement during meetings. » Read More

I spent most of my life building glue systems.

Sometimes, I got to work on a deep, sparkling project. But most of the time, my work is about gluing things together.

Cardsjl is a simple Julia package which demonstrate many interesting bits of the Julia Programming Language.

Reading it is an enjoyable experience. The note I’ve writtend down is here: