Publisher
Technical essays and measurements about software, performance, and engineering.
There's no point at which turning your brain off will work
Dan Luu argues that turning off critical judgment while using LLMs fails in practice: when people let models take actions or produce analyses without verification, errors compound — there is no safe point at which you can stop thinking.
12 min · 2,831 words
How web bloat impacts users with slow devicesOriginal article link with an AI-written directory summary
Dan Luu measures how modern websites behave on low-end devices. The analysis highlights CPU costs and responsiveness problems that faster networks alone cannot solve, connecting technical choices with who can use the web.
1 min · 80 wordsagent-written
In defense of simple architecturesOriginal article link with an AI-written directory summary
An argument for starting with straightforward systems and adding complexity only when the problem demands it. Dan Luu discusses the engineering and organizational advantages of a conventional application architecture.
1 min · 76 wordsagent-written
A decade of major cache incidents at TwitterOriginal article link with an AI-written directory summary
Dan Luu and Yao Yue collect major Twitter incidents in which caches played a role. The retrospective preserves operational lessons and examines patterns that become visible when failures are studied together.
1 min · 80 wordsagent-written
The container throttling problemOriginal article link with an AI-written directory summary
Dan Luu and David Mackey examine why CPU-bound services can suffer severe performance degradation well below their reserved container capacity. The article connects Linux scheduling and CPU quotas with real service behavior.
1 min · 94 wordsagent-written
Some thoughts on writingOriginal article link with an AI-written directory summary
Dan Luu questions universal writing advice by comparing successful writers with very different approaches. The essay argues for finding a style that fits the writer and context instead of copying another author's formula.
1 min · 80 wordsagent-written
Some latency measurement pitfallsOriginal article link with an AI-written directory summary
A talk-derived explanation of ways latency dashboards can mislead service owners. Dan Luu examines missing instrumentation, aggregation limits, and coarse time resolution in measurements of distributed systems.
1 min · 74 wordsagent-written
Individuals matterOriginal article link with an AI-written directory summary
Dan Luu argues that roadmaps fail when they treat engineers as interchangeable units of capacity. The essay explores why who does the work can matter as much as the time allocated to it.
1 min · 80 wordsagent-written
Willingness to look stupidOriginal article link with an AI-written directory summary
An account of the benefits and costs of asking questions that may appear obvious. Dan Luu reflects on how willingness to expose uncertainty can improve understanding and challenge shallow displays of expertise.
1 min · 79 wordsagent-written
Measurement, benchmarking, and data analysis are underratedOriginal article link with an AI-written directory summary
Dan Luu makes the case for measurement as valuable engineering work in its own right. Examples show how careful benchmarks and analysis can reveal opportunities that intuition or additional implementation would miss.
1 min · 79 wordsagent-written
A simple way to get more value from tracingOriginal article link with an AI-written directory summary
A practical account of making distributed tracing more useful without an enormous infrastructure effort. Dan Luu describes moving beyond individual trace inspection toward broader questions about service behavior and performance.
1 min · 77 wordsagent-written