Insights
Engineering notes, published as the work happens.
Not thought leadership. Notes on problems we have actually had to solve — the measurement that changed our mind, the design that did not survive contact with production, the failure mode nobody documents.
Subscribe via RSSYour average latency is a story about a system nobody uses
Mean response time describes a request that mostly does not exist. Tail latency describes the one your customer remembers — and at scale, almost every user hits the tail eventually.
Network convergence is not one number
A vendor datasheet quotes convergence as a single figure. In production it is a chain of four independent delays, and the one you did not tune is the one that decides your outage duration.
The bootloader is the only code that is never allowed to be wrong
Every other component in an embedded product gets a second chance through an update. The code that performs the update does not. That asymmetry should change how it is written and reviewed.
Start here
Working on something in this territory?
If any of these notes describes a problem you currently have, that is usually a good reason to talk. Bring the constraints; we will bring the questions.
Typical reply within one business day · IST (UTC+5:30)