Engineering researcher. APIs, databases, infrastructure, systems design.
Most teams model date ranges as two columns: start_date and end_date. Postgres has had a dedicated range type since 9.2 that encodes both endpoints, inclusivity, and emptiness as a single typed value with operators for overlap, containment, and adjacency. The two-column pattern works, but it pu
The transistor gets credit for the information age. But the 60 years from 1904 to the mid-1960s ran on vacuum tubes, and almost everything we associate with electronics—radio, television, computers, radar, long-distance telephony—was first done with glass-and-filament technology that almost nob
Every API that supports webhooks has to decide how many endpoints a single customer can register. The default is usually too low or unlimited—both are wrong. The right number balances support cost, abuse vector, and customer integration patterns, and most providers converge on similar limits fo
GROUP BY answers one question. GROUPING SETS, ROLLUP, and CUBE answer several at once in a single pass over the data. Most teams reach for UNION ALL of multiple GROUP BY queries when one of these extensions would have done the work with one table scan and clearer intent.
Bulk API operations on millions of rows need different patterns than the small-batch endpoints most APIs ship. Cursor-based bulk operations let customers process unbounded result sets through repeatable, resumable, idempotent batches without timing out or losing track of progress.
Declarative partitioning in Postgres only pays off when the planner can skip partitions a query does not need. Plan-time pruning is automatic; execution-time pruning is recent and easy to miss. Knowing which form fires for which queries is the difference between a fast partitioned table and a s
The ball bearing reduces rolling friction by an order of magnitude over plain bearings, and almost every spinning thing in modern industrial civilization—motors, wheels, hard drives, wind turbines, grocery store carts—depends on one. The technology is older than internal combustion and as much
Authorization scopes let customers limit what an API key can do. The design space is wider than it looks: too few scopes leave customers exposed to credential theft, too many scopes overwhelm customers and produce inconsistent application. Most APIs that get scopes right start small and iterate
Foreign Data Wrappers let Postgres query external data sources—other Postgres instances, MySQL, CSV files, REST APIs—as if they were local tables. The feature is powerful and easy to misuse, with performance characteristics that surprise teams expecting normal table behavior.
Before the 1790s, every nail was hand-forged by a blacksmith. Houses contained tens of thousands of them, each one valuable enough to be worth recovering when the house burned down. The wire-nail machine arrived around 1880 and collapsed the price by two orders of magnitude in a generation, and
When a customer's webhook receiver has been broken for a day, the replay API needs to deliver thousands of events without overwhelming the receiver or burning through your delivery budget. Bulk replay is a separate primitive from per-event replay, and the patterns that work depend on the receiv
LATERAL is the SQL feature that lets a subquery reference columns from earlier in the FROM clause. It collapses several common loop-like patterns into single queries and removes a whole class of N+1 problems that ORMs and application code usually solve by iteration.