01 August 2026

17 new posts across 8 engineering blogs — 7 worth your time.

One week of Engineering Reading, 7 stories, as published.

Post-quantum authentication to origins is now supported

Cloudflare's Authenticated Origin Pulls and Custom Origin Trust Store now support post-quantum authentication. The post explains how to configure fully post-quantum secure mutually authenticated TLS connections to origin servers. It also dives into the engineering details of how Cloudflare built this and how it fits into their overall post-quantum migration.

Why it matters — This article provides engineering details on implementing post-quantum authentication for secure TLS connections to origin servers.

Cloudflare · 01 August 2026 · Read the original →

Data & ML

GenRec: Towards LLM-Native Recommendation at Netflix

Recommendations are central to the Netflix experience, with current production models relying on thousands of hand-crafted features and specialized architectures for sequence modeling, feature interactions, and multi-task objectives. This existing stack has evolved over many years to support diverse content. The post introduces GenRec, signaling a move towards LLM-native recommendation at Netflix.

Why it matters — It outlines the evolution of Netflix's core recommendation system and its shift towards LLM-native approaches.

Netflix · 01 August 2026 · Read the original →

Performance

Don’t stop early: Case-folding source code at memory speed

For accurate text matching in search engines, regex, and case-insensitive contexts, a canonical form called case folding is necessary to erase case distinctions. This ensures that strings differing only in case compare as equal, such as "café" and "CAFÉ." The post discusses achieving this case folding at memory speed for source code.

Why it matters — This article details the technical problem and solution for high-performance case folding in text matching systems like search engines.

GitHub · 01 August 2026 · Read the original →

Data & ML

Indexing the Data Lake for Online Point Queries

Companies such as Spotify require vast quantities of data to be accessible at low latency for their online services. This post focuses on the specific challenge of indexing the data lake to support online point queries. It aims to address how to efficiently retrieve individual data points from large datasets with minimal delay.

Why it matters — It addresses the critical problem of making vast data lake quantities accessible at low latency for online point queries.

Spotify · 01 August 2026 · Read the original →

Data & ML

Eval-driven development: Lessons from evaluating GenAI at scale

Airbnb teams build trustworthy Generative AI products by treating evaluation as a first-class engineering discipline, rather than an afterthought. Generative AI breaks many traditional assumptions, necessitating this approach. The post shares lessons from evaluating GenAI at scale to ensure product reliability.

Airbnb · 01 August 2026 · Read the original →

Data & ML

Kafka App? There’s a Skill for That

Etsy manages over 100 million unique listings from 5.6 million sellers, lacking a standard product catalog. When shoppers search, models must infer meaning and listing content using data from user visits and Etsy listings. Kafka is utilized in this process to enable these models to understand unique and creative items.

Why it matters — It explains how Etsy uses Kafka and data models to infer meaning from unique listings for complex search queries.

Etsy · 01 August 2026 · Read the original →

Data & ML

Pinner Progression: Better Use-Case Representation Driving Weekly Active User Growth at Pinterest

Pinterest's mission is to help people discover, and this post, part 1 of 2, focuses on "Pinner Progression." It aims to achieve better use-case representation to drive weekly active user growth. The article involves contributions from personalization, user understanding, and data science teams, suggesting a deep dive into how these areas contribute to user engagement.

Why it matters — This post details how better use-case representation in personalization and user understanding drives weekly active user growth.

Pinterest · 01 August 2026 · Read the original →

Lots of expensive ways to do a key-value lookup this week.

7 stories, every Saturday

Published here every week. Follow by RSS to get it as it lands.

Subscribe by RSS All issues

Next issue →