For below profiles/videos/links, from Netflix linked profiles

• mostly derived from Philip’s sharing https://www.linkedin.com/in/philfish/
• and Mick’s sharing https://www.linkedin.com/in/mdreeling/
• some of the architectures are out of dated, but still valid for the logistics behind tradeoff and decision

@Xin:

• over all the talks here, my feeling is, looking at experiences of every speaker or author, it’s better to stick to a great place for a long time, and grow domain knowledge from there
• think even harder, I watched again Steve Jobs’ Stanford Commencement Address

Software process

https://netflixtechblog.com/full-cycle-developers-at-netflix-a08c31f83249

• Operating What You Build. well same here in Amazon lol
• Scaling Through Developer Tools. well, natural and instintive to me, maybe staying at Amazon too long…
• Full Cycle Developers

[worth-watching-rank-high] Edge Tech Architecture video

https://www.youtube.com/watch?v=5ju4W9KAzcY

• Ownership:signup, discovery, playback
• detailed in below Satyajit Thadeshwar’s talk

Playback data systems team and job overview

https://www.slideshare.net/SuudhanRangarajan/netflix-playback-data-systems-team-and-job-overview

• Justin Becker, ~10 years Netflix tenure https://www.linkedin.com/in/jbeck449/

• region-based, like AWS
• Cassandra, AWS-SQS, Kafka, Hive

Phillipa Avery: What Inspires you at Netflix?

https://www.youtube.com/watch?v=PYDk20IpulA

Phillipa Avery from the Playback Security Team

• https://www.linkedin.com/in/phillipa-avery-138ba99/
• 6.6 year Netflix tenure to now

[worth-watching-rank-medium] [Video] Scaling Patterns for Netflix’s Edge

https://netflix.github.io/mantis/

• Mantis is a platform to build an ecosystem of realtime stream processing applications.
  • stream processing As-a-Service
    • https://netflix.github.io/mantis/gettingstarted/concepts/
    • https://netflix.github.io/mantis/internals/infrastructure/
• @Xin: I think it’s the same idea of cloudwatch agent
  • AWS for the wi…n… ?

[worth-watching-rank-low] [Video] Netflix Play API

• video: https://www.youtube.com/watch?v=6oPj-DW09DU
• Justin Ryan https://www.linkedin.com/in/justin-charles-ryan/

@Xin: Basically this video is some high level design tradeoffs, not tech strong.

@Xin: Always aligned with design goal, before talking about decisions

• Identity: basically micro-service tradeoff, transitioning to a unified api across the whole company. Just like AWS-Identity team.
• type 1/2 decisions: from JeffB…hahha… one-way-door or two-way-door
• types: coupling level, asynchronous, data architecture
• language coupling: good point, “coupling encouraging sticking with Java”

[worth-watching-rank-high] User & Device Identity for Microservices @ Netflix Scale

• video https://www.infoq.com/presentations/netflix-user-identity/
• Satyajit Thadeshwar, 10 years veteran at Netflix https://www.linkedin.com/in/satyajit-thadeshwar/

Notes

• identity service: 2M request/second
• move authentication to Edge(Zuul), with 5% request redirect to new service
• passport: abstraction layer of identity info across Netflix internal services (header, user, device, integrity)
  • passport action: response to client request. use case: login, or switch user
• extensible identity model: flexible extra parameters

[worth-watching-rank-high] A/B Testing

https://hbr.org/2020/03/productive-innovation#avoid-the-pitfalls-of-a-b-testing

Netflix Playback Access Team

https://www.slideshare.net/kcasella/netflix-playback-access-team

• Karen Casella https://www.linkedin.com/in/kcasella/

SDE example - Playback Data Systems

Tech stack :

• Handling large datasets using distributed data system techniques such as eventual consistency, partitioning and caching
• Experience using Pub-sub (Kafka) and stream processing (Spark/Flink/Cloud Dataflow/etc) systems
• Utilizing NoSQL data stores (Cassandra, Elasticsearch, HBase, Redis etc.) in modeling and evolving domain-specific application data.
• Familiarity with Cloud based architectures (AWS/Azure/GCP)