N
Netflix Tech Blog
netflixtechblog.com · Company blog · on HN since 2020
Netflix engineers write about architecture, performance debugging, video encoding, and resilience on the company's engineering blog.
What HN saysCommenters cite it as a good example of a tech blog that shares real information rather than marketing, and list it among strong engineering blogs.
363HN people shared it
157linked it in comments
475stories submitted
9.3Kstory points
2 months agolast shared
2020people per year2026
Most-shared posts
- 1Homepage13 people · first shared 2020
- 2The Netflix Cosmos Platform: Orchestrated Functions as a Microservice8 people · 42 points · first shared 2021 · 1 comments on HN
- 3Hexagonal Architecture at Netflix8 people · 16 points · first shared 2020 · 1 comments on HN
- 4The Case of the Extra 40ms7 people · 783 points · first shared 2020 · 197 comments on HN
- 5Linux Performance Analysis (2015)7 people · 188 points · first shared 2021 · 40 comments on HN
- 6Seamlessly Swapping the API back end of the Netflix Android app7 people · 65 points · first shared 2020 · 8 comments on HN
- 7Netflix Scales Its API with GraphQL Federation (Part 1)7 people · 14 points · first shared 2020 · 2 comments on HN
- 8Netflix is now doing per-shot encoding for UHD content6 people · 355 points · first shared 2020 · 196 comments on HN
- 9Open Sourcing the Netflix Domain Graph Service Framework GraphQL for Spring Boot6 people · 18 points · first shared 2021 · discussion on HN
- 10Netflix Drive: A file folder interface for Netflix Cloud Services6 people · 13 points · first shared 2021 · 1 comments on HN
- 11AVIF for Next-Generation Image Coding5 people · 195 points · first shared 2020 · 79 comments on HN
- 12How Netflix uses Druid for realtime insights5 people · 109 points · first shared 2020 · 59 comments on HN
- 13Building Netflix's Distributed Tracing Infrastructure5 people · 106 points · first shared 2020 · 17 comments on HN
- 14Computational Causal Inference at Netflix5 people · 14 points · first shared 2020 · discussion on HN
- 15Telltale: Netflix Application Monitoring Simplified5 people · 10 points · first shared 2020 · discussion on HN
What HN people wrote when they linked it
colechristensen · 2025-10-21
There's a tone issue.
Marketing should be marketing and clearly so. Tech blogs are about sharing information with the community (Netflix Tech blog is a good example) NOT selling something. Marketing masquerading as a tech blog is offputting to a lot of people. People don't like being fooled with embedded advertising and putting ad copy into such pieces is at best annoying.
https://netflixtechblog.com/
jedberg · 2026-05-09
We were doing multi-AZ and multi-region failover at Netflix all the way back in 2011:
https://netflixtechblog.com/the-netflix-simian-army-16e57fbab116
vivzkestrel · 2026-03-04
- can someone kindly explain why there are 2 websites that all claim to be netflix tech blog?
- website 1 https://netflixtechblog.medium.com/
- website 2 https://netflixtechblog.com/
brianmz · 2026-01-27
Multiple high quality company engineering blogs including https://netflixtechblog.com/, https://www.uber.com/en-US/blog/engineering/, DoorDash etc.
https://techtalksweekly.io/ - new software dev conference talks and podcasts
https://ciechanow.ski/ - interactive articles about science and engineering
https://jvns.ca/ - great technical content overall
This is a great thread btw.
tester756 · 2025-12-23
bzGoRust · 2025-12-23
SerCe · 2025-12-15
Great article! Just yesterday I watched a Devoxx talk by Andrei Pangin [1], the creator of async-profiler where I learned about the new heatmap support. To many folks it might not sound that exciting, until you realise that these heatmaps make it much easier to see patterns over time. If you’re interested there’s a solid blog post [2] from Netflix that walks through the format and why it can be incredibly useful.
[1]: https://www.youtube.com/watch?v=u7-S-Hn-7Do
[2]: https://netflixtechblog.com/netflix-flamescope-a57ca19d47bb
bmcahren · 2025-08-22
This was a good read and great work. Can't wait to see the performance tests.
Your write up connected some early knowledge from when I was 11 where I was trying to set up a database/backend and was finding lots of cgi-bin online. I realize now those were spinning up new processes with each request https://en.wikipedia.org/wiki/Common_Gateway_Interface
I remember when sendfile became available for my large gaming forum with dozens of TB of demo downloads. That alone was huge for concurrency.
I thought I had swore off this type of engineering but between this, the Netflix case of extra 40ms and the GTA 5 70% load time reduction maybe there is a lot more impactful work to be done.
https://netflixtechblog.com/life-of-a-netflix-partner-engineer-the-cas…
https://nee.lv/2021/02/28/How-I-cut-GTA-Online-loading-times-by-70/
dgs_sgd · 2025-08-10
Awesome work! I would suggest adding Netflix’s blog [1]. It’s very high quality imo.
[1] https://netflixtechblog.com/
vtuulos · 2025-07-17
I don't know if it's a coincidence but we just released a major new feature in Metaflow a few days ago - composing flows with custom decorators: https://docs.metaflow.org/metaflow/composing-flows/introduction
A big deal is that they get packaged automatically for remote execution. And you can attach them on the command line without touching code, which makes it easy to build pipelines with pluggable functionality - think e.g. switching an LLM provider on the fly.
If you haven't looked into Metaflow recently, configuration management is another big feature that was contributed by the team at Netflix: https://netflixtechblog.com/introducing-configurable-metaflow-d2fb8e9b…
Many folks love the new native support for uv too: https://docs.metaflow.org/scaling/dependencies/uv
I'm happy to answer any questions here
gregors · 2024-12-19
I'd love to know how much faster we've gotten since this was released (now unmaintained)
https://netflixtechblog.com/fast-json-api-serialization-with-ruby-on-r…
I really thought this pure ruby implementation was neat, though I never actually used it in production. It's been long since abandoned.
I'm curious in general about the status of pure ruby implementations. It looks like they killed json_pure? If so, that's a bummer. Does anyone know the details? By far the most interesting part of the posts to me are the Ruby optimizations not so much the C ones.
jerf · 2024-09-19
For anyone looking for a TL;DR, I'd suggest starting at https://netflixtechblog.com/introducing-netflixs-key-value-data-abstra… , which HN is truncating so you can't see it but I've directly linked to a later section in the post with a #. Up to that point it's basically "a networked HashMap<String, SortedMap<Bytes, Bytes>>". But the ability to return partial results based on a timeout with a pagination token is somewhat unusual and the next section called "Signaling" is at least worth a look.
throw67564 · 2024-05-09
This article seems to be relevant
>This blog post sheds light on the groundbreaking series of A/B tests Netflix did which resulted in increased member engagement. Our goals were the following:
>Identify artwork that enabled members to find a story they wanted to watch faster [0]
Although it doesn't say that they give different thumbnails to different users, just that they tested different thumbnails for overall performance using A/B testing. So I'd like a source for customized per-user thumbnails. Although I only skimmed the article so I might have missed it.
[0] https://netflixtechblog.com/selecting-the-best-artwork-for-videos-thro…
j2kun · 2024-01-13
This article directly contradicts an article written by Netflix engineers Xavier Amatriain and Justin Basilico, who were in charge of putting the BellKor algorithm in production: https://netflixtechblog.com/netflix-recommendations-beyond-the-5-stars…
> But once we overcame those challenges, we put the two algorithms into production, where they are still used as part of our recommendation engine.
devta · 2023-12-26
maxpert · 2022-10-17
While I love ingenuity of developers at Hasura (because I've been personally through these scaling challenges), I always get a gag reaction with GraphQL. I've honestly tried hard to digest it, and I can tell you at large scale where single DB won't cut it, you would either need to develop a large federation layer like [Netflix](https://netflixtechblog.com/how-netflix-scales-its-api-with-graphql-fe… or just rip-it out. Streaming might elevate the problem a little, but I real problem still lurks under the hood. The fact that front-end community wants to fit everything under GraphQL bothers me, because every backend developer knows that a single tool/technology is usually not the best tool for solving all problems of your life. Remember the golden words, THERE IS NO SILVER BULLET!
frakt0x90 · 2022-09-08
Agreed the article doesn't make it clear. I know companies like Netflix have built knowledge graphs of their products which they create embeddings of to feed to ML models which helps them make better recommendations. NASA also built one to ease search of safety information when building new things.
https://netflixtechblog.com/supporting-content-decision-makers-with-ma…
https://neo4j.com/blog/nasa-critical-data-knowledge-graph/
carride · 2022-04-22
Perhaps they promote their tech achievements better than others https://netflixtechblog.com/
BlueTemplar · 2022-01-09
I would assume that's quite a hard project, considering that even for Netflix it took two decades of research and world-class AI competition(s?) with a million dollar prize(s?) :
https://netflixtechblog.com/netflix-recommendations-beyond-the-5-stars…
(Note that was for the much more USA-focused DVD by mail Netflix, so the current problem might be even harder to figure out !)
vjrajto · 2022-01-01
darkwizard42 · 2021-08-26
I believe Netflix has talked about thumbnail experimentation at length in some of their engineering blogs [1]. To me, it seems perhaps Netflix has figured out a running theme in the author's viewing habits... perhaps the scandalous, soapy stuff is what they keep clicking.
Same kind of reaction I get when a guy friend tells me they keep getting ads for women's bikinis or women's underwear in their Instagram ads. It isn't sexualization of Instagram... it is a reflection of your interaction with the platform!
[1: https://netflixtechblog.com/selecting-the-best-artwork-for-videos-thro…
cogman10 · 2021-05-17
Last I heard, netflix has studios send them progressive video and lets them deal with de-interlacing.
AFAIK, for enthusiasts, the best deinterlacer is "QTGMC" http://avisynth.nl/index.php/QTGMC
For stuff shot like post 2015, I don't think anyone is interlacing anymore. Everything is progressive.
https://netflixtechblog.com/high-quality-video-encoding-at-scale-d159d…
manca · 2021-04-26
This is a known and hard problem. A lot of companies are trying to internally do some sort of analysis of their own content libraries and intelligently tune their encoders. Netflix is definitely the most known for their tech related to video analysis and introduction of VMAF[1], but other metrics also exist which enable you to compare the original/master and the encoded variant (PSNR, SSIM, etc). Bottom line is that you need a lot of trial and error and fitting different curves on your bitrate/quality graphs and often times what metrics consider to be good quality human visual system doesn't agree 100% with. It's a very interesting problem nonetheless, and I recommend [2] if you want to learn more.
[1] https://en.wikipedia.org/wiki/Video_Multimethod_Assessment_Fusion
[2] https://netflixtechblog.com/toward-a-practical-perceptual-video-qualit…
knowhy · 2021-01-13
I don't see the benefit either. I believe Netflix made this idea popular with sidecar containers as a way to add
> “non-intrusive platform capabilities” [0]
to container stacks.
I understand that the idea might be popular among developers as it is easier to just add a database container to your stack rather than dealing with the db admin. I don't know if Netflix ever recommended databases as sidecar containers. But I have seen it in the wild where dev followed the Netflix model. I sometimes hear people arguing that they have to manage containers anyway so it would be less overhead to manage the db as a container as well.
0: https://netflixtechblog.com/prana-a-sidecar-for-your-netflix-paas-base…
clouddrover · 2021-01-10
nixgeek · 2020-09-28
They write about some of the interesting engineering work on various blogs.
https://netflixtechblog.com/
adwi · 2020-09-26
Netflix are masters at video compression, great reads and insights if you’re so inclined:
https://netflixtechblog.com/tagged/video-encoding
theandrewbailey · 2020-06-23
It's based on the next-gen AV1 video codec. I've tried it out, and it's amazing.
https://netflixtechblog.com/avif-for-next-generation-image-coding-b1d7…
https://github.com/AOMediaCodec/libavif
pas · 2020-02-22
Netflix loves Cassandra, right? [0][1] So could someone describe why it wasn't a great fit for Uber? How come it was easier to invent the wheel in Go compared to cobbling together something with Cassandra/ES/Kafka (or other Java gadgets from the Hadoop ecosystem)?
[0]: https://netflixtechblog.com/scaling-time-series-data-storage-part-i-ec…
[1]: https://www.datastax.com/resources/video/cassandra-netflix-and-version…
minimaxir · 2020-02-06
Netflix still does a ton of statistical experiments. A good one is video thumbnail selection.
2016 blog post: https://netflixtechblog.com/selecting-the-best-artwork-for-videos-thro…
2018 presentation: https://www.youtube.com/watch?v=UjQMEjkrUGo