Velnoxis

Velnoxis

Long-form technical notes, written by operators for operators.

Compliance

Data Residency Basics for Global Teams

June 24, 2026

Data residency questions arrive late in most projects, usually phrased as: can you just keep EU customer data in the EU? The honest answer is that 'just' is doing enormous work in that sentence, because residency is not a hosting decision but an architecture decision.

The trap is assuming a region pin on your primary database solves it. Backups, search indexes, analytics exports, email logs, and crash reports all replicate data, often across borders by default. Auditing those paths is most of the actual work.

Continue reading →

Operations

Reading Latency Percentiles Without Fooling Yourself

April 11, 2026

Averages conceal what percentiles reveal: the request mix's tail. A service averaging fifty milliseconds can still be sending one in twenty users through a two-second odyssey, and those users - the ones hitting the cold paths, the expired sessions, the overloaded shard - are the ones filing the tick…

Engineering

A Field Guide to Graceful Degradation

May 27, 2026

Complex platforms require predetermined shedding tiers when capacity degrades. Personalized feeds should drop before transactional checkout; suggestion dropdowns should degrade before primary queries; media thumbnails should omit before raw assets. Documenting and enforcing these boundaries via circ…

Security

Understanding TLS 1.3 Session Resumption

May 2, 2026

Shortening the standard handshake to a solitary round trip was an achievement, but session resumption delivers the true speed advantage for returning traffic. Transmitting data right away via pre-shared keys spares high-latency cellular connections the drag of multi-step cryptographic exchanges.…

Operations

Zero-Downtime Deployments Without the Drama

May 19, 2026

Popular engineering lore depicts seamless releases through elaborate blue-green switches and instantaneous traffic flips. Production reality centers on disciplined fundamentals: synthetic probes verifying realistic workloads, graceful connection draining on retiring nodes, and database schema mutati…

More reading

About us

Our contributors have spent years on-call for large platforms. This site collects the playbooks, postmortems and reference material we wish someone had handed us earlier.

More about the project →