<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>CoreSignal — Engineering signal. Zero noise.</title><link>https://coresignal.pages.dev/</link><description>A daily, curated engineering briefing.</description><language>en-ie</language><item><title>Meta Andromeda redesigns ads retrieval around models and hardware</title><link>https://coresignal.pages.dev/articles/c4d409a1-3fbc-4b98-bdbc-8517853a1302/</link><guid isPermaLink="true">https://coresignal.pages.dev/articles/c4d409a1-3fbc-4b98-bdbc-8517853a1302/</guid><pubDate>Sun, 15 Nov 2026 22:47:58 GMT</pubDate><description>Meta’s next-generation Andromeda brings a more capable model to the first stage of ad selection by designing the model, execution pipeline, and GPU use together. The goal is to evaluate a much larger pool of ads without spending proportionally more compute or exceeding the existing latency budget.</description><category>ai-ml</category><source url="https://engineering.fb.com/2026/10/28/ml-applications/meta-andromeda-next-gen-ads-retrieval-gpu-co-design/">Meta Engineering</source></item><item><title>Disaster recovery readiness takes more than a scheduled failover</title><link>https://coresignal.pages.dev/articles/e67cd020-5525-4ca7-bbf9-226930d6e660/</link><guid isPermaLink="true">https://coresignal.pages.dev/articles/e67cd020-5525-4ca7-bbf9-226930d6e660/</guid><pubDate>Sun, 15 Nov 2026 22:47:58 GMT</pubDate><description>A disaster recovery plan is a commitment on paper until teams test it. Progressively more demanding exercises reveal different gaps, from unclear ownership to failures that were not anticipated in the plan.</description><category>distributed-systems</category><source url="https://aws.amazon.com/blogs/architecture/validating-your-dr-commitment-from-controlled-test-to-real-world-readiness/">AWS Architecture</source></item><item><title>Making Production Database Traffic Useful for Safer Testing</title><link>https://coresignal.pages.dev/articles/a8a576fa-e5ce-4cd3-a083-b3f256d5d98f/</link><guid isPermaLink="true">https://coresignal.pages.dev/articles/a8a576fa-e5ce-4cd3-a083-b3f256d5d98f/</guid><pubDate>Sun, 15 Nov 2026 22:47:58 GMT</pubDate><description>Airbnb built a database traffic capture and replay system around a proxy that already sits between applications and databases. It uses real queries, rather than relying solely on synthetic workloads, to test capacity, compare database behavior and investigate performance problems offline.</description><category>cloud-infrastructure</category><source url="https://medium.com/airbnb-engineering/beyond-synthetic-testing-capturing-and-replaying-real-database-workloads-at-airbnb-cea7ee9b1ab2?source=rss----53c7c27702d5---4">Airbnb Engineering</source></item><item><title>TLX gives ML engineers more control over GPU execution</title><link>https://coresignal.pages.dev/articles/4a57f1c1-54a6-4271-aacb-a711bd636dc7/</link><guid isPermaLink="true">https://coresignal.pages.dev/articles/4a57f1c1-54a6-4271-aacb-a711bd636dc7/</guid><pubDate>Sun, 15 Nov 2026 22:47:58 GMT</pubDate><description>Triton Low-Level Extensions, or TLX, extends Triton so engineers can organize work across modern GPU execution resources without abandoning its productive programming model.</description><category>ai-ml</category><source url="https://engineering.fb.com/2026/11/10/open-source/tlx-a-hardware-native-gpu-compiler-for-production-ml/">Meta Engineering</source></item><item><title>Making older time-series data cheaper without changing the query interface</title><link>https://coresignal.pages.dev/articles/58827444-e26a-484f-9a19-bb6894f54f01/</link><guid isPermaLink="true">https://coresignal.pages.dev/articles/58827444-e26a-484f-9a19-bb6894f54f01/</guid><pubDate>Sun, 15 Nov 2026 21:04:35 GMT</pubDate><description>Netflix’s TimeSeries platform separates frequently accessed recent events from older data, moving the latter to object storage while keeping a consistent query interface. Its first approach streams events into a cold tier and reorganizes them for efficient reads. The source also introduces a Cassandra-native approach that uses backups already in object storage.</description><category>distributed-systems</category><source url="https://netflixtechblog.com/timeseries-tiered-storage-journey-kafka-flink-streams-to-native-cassandra-cold-reads-e59d597c9d60?source=rss----2615bd06b42e---4">Netflix TechBlog</source></item></channel></rss>