Anurag’s Read

across 12 engineering blogs:

, , , , , , , , , , , .
7.0

334 of 565 posts

Reading list

Netflix27

9.2Behind the Scenes: Evolving Netflix's Ads Event Pipeline for Live — Part II
8.8TimeSeries Tiered Storage Journey: Kafka/Flink Streams to Native Cassandra Cold Reads
8.8Netflix's Distributed Counter Abstraction
8.6Netflix Stratum Media Processing: Automated Container Right-Sizing

Canva6

9.2Canva incident report: API Gateway outage
9.1Session revocations at scale
8.7Worker Backpressure (Part 1)
8.6The foundations of Canva's continuous data platform with Snowpipe Streaming

AWS28

9.1Using load shedding to avoid overload
9.0Just make it scale: An Aurora DSQL story
8.7S3 Files and the changing face of S3
8.7Avoiding insurmountable queue backlogs

PlanetScale42

9.0Designing a reliable unplanned failover
8.8Larger than RAM Vector Indexes for Relational Databases
8.7The lifecycle of a sharded Postgres query
8.6What gets TOASTed in Postgres

Cloudflare47

9.0Cutting our inference bill by 68% by being obsessive about optimization
9.0Cloudflare outage on November 18, 2025
9.0Migrating billions of records: moving our active DNS database while it's in use
9.0Building Vectorize, a distributed vector database, on Cloudflare's Developer Platform

Discord12

9.0You've Got (Too Much) Mail: Behind the Scenes of the 3/25/26 Voice Outage
9.0How Discord Indexes Trillions of Messages
8.9How We Moved Discord Voice to the Edge
8.7Tracing Discord's Elixir Systems (Without Melting Everything)

Stripe13

9.0Selective Test Execution at Stripe: Fast CI for a 50M-line Ruby monorepo
8.7How Stripe uses graph search and state machines to auto-remediate a global database fleet
8.6Building a data plane from scratch: Stripe's own high-performance distributed proxy
8.5Formatting an entire 25 million line codebase overnight: the rubyfmt story

Slack13

9.0Build better software to build software better
8.6From SSH to REST: A Security-Driven Modernization of Slack's EMR Data Pipelines
8.6Deploy Safety: Reducing customer impact from change
8.5Slack AI: The Path to Multi-Cloud

Uber75

8.9How Uber Protects Against Retry Storms
8.8Translations at Scale: Memory Optimization Techniques That Kept Uber's P99 Under 1ms
8.8How Uber Conquered Database Overload: The Journey from Static Rate-Limiting to Intelligent Load Management
8.7How Uber Migrated ETL from Hive to Spark SQL

Linear6

8.8Rebuilding Linear's delta sync read path
8.6AI coding has made CI a bottleneck, so we reworked ours to keep up
8.2Post mortem on Linear security incident on March 24th, 2026
8.1Styling Linear for the future with StyleX

GitHub18

8.6From 2,500 requests to 24: Rebuilding star history without the stargazers
8.6The uphill climb of making diff lines performant
8.5Improving site performance by shipping more CSS
8.4From latency to instant: Modernizing GitHub Issues navigation performance

Meta47

8.6ZGateway: Learnings from Putting a Proxy in Front of ZippyDB
8.6Strobelight: A profiling service built on open source technology
8.4Translating Java to Kotlin at Scale
8.3Adopting AV1 for Real-Time Communication (RTC) at Scale