Omkaar Kamath

Intuition behind Distributed Training: Busy Engineer Edition

Distributed training seems complex at first with all the deep learning jargon but is pretty intuitive... let's understand it from scratch

How Yelp Switched to a Near-Real-Time Search Achitecture

How Dropbox saved $1.7M using a new machine learning classifier system

How Discord indexes 100B+ messages without breaking their bank

How Uber improved Cassandra reliability by 1000%

How Github's load balancer handles 5B+ requests / day

How Doordash Cut Elasticache Costs by 50%

How Doordash scaled up their feature store to 20M reads per second (Part 1)

Scaling Discord's message datastore to 1,000,000,000,000 messages

How Twitter creates sortable and unique IDs for 200B Tweets per year