Instead of an introduction, I am linking to my short blog post based on a coffee chat: A Manager’s Incentives.
Software Engineering ⚙️
Detecting regressions, especially non-deterministic regressions, is about finding the tricky balance between false positives, which will reduce the trust in the detection, and false negatives, which will release regressions to customers. Netflix’s TVUI team adopted statistical analysis (anomaly and changepoint detection) to reduce both falses.
Not just JS tools but CSS tools are now written in Rust.
Throughout my career, I was disappointed that the standups eventually turned into mere status updates and, worse yet, daily guilt trips. The engineers felt compelled to show that they were productive the day before. This StackOverflow thread showed me an alternative; teams focus on what they need from each other, not what they did or will do.
This article reminds me of my past experiences where the more mature, successful companies (Netflix, especially) had far more engineers interested in the companies’ businesses than the failed startups had. Curiosity and respect for other functions and customers will make us better engineers and more effective teams.
This first-hand account of looking for a new role as a senior engineer explains why he decided to leave and how he looked for the next opportunity. I especially found the idea of treating compensation as a trailing indicator enlightening.
A group called the Data & Trust Alliance created a guideline to safeguard against bias when algorithms are used to make HR decisions, such as hiring and promotions. The fact that this guideline exists shows that employers are already using algorithms to make those decisions. Companies like Meta, Starbucks, and Pfizer are members of the Alliance.