Receive the latest technical articles and release notes in your inbox.
Retention at 3 days with a refresh window reaching back 7 means the aggregate keeps no more history than the raw table. Here’s the invariant that fixes it.
By John Shi
October 23rd, 2026
Five refresh policy shapes for solar and wind SCADA: frequent, daily, hierarchical, retention-aware, and manual. Pick the one that fits each job.
By Damaso Sanoja
October 7th, 2026
Reshape a wide oilfield SCADA table into a narrow hypertable plus tag metadata, in place, with dashboards running and history left where it is.
By Damaso Sanoja
September 30th, 2026
One Postgres table with two indexes beat every published RAG pipeline on MuSiQue. Every architectural addition is a bet the model stays weak.
By Matvey Arye
September 30th, 2026
Built for industrial, energy, and robotics systems. Run high-ingest telemetry and real-time analytics on the Postgres you already know.
No fact extraction, no graphs, no memory manager. One Postgres table and three search tools scored F1 0.665 on LoCoMo.
By Matvey Arye
September 29th, 2026
41 experiments in 6 days took LoCoMo memory from F1 0.392 to 0.666. Claude wrote the code; the human caught the metrics that were lying.
By Matvey Arye
September 28th, 2026
A 20-byte sensor reading costs 95.6 on disk. Measure bytes per row, project 631 billion rows, and test your ingest ceiling before the schema review.
By John Shi
September 25th, 2026
Composite indexes charge every insert and pay back only some queries. Four steps to price both sides and audit the indexes you already run.
By John Shi
September 18th, 2026
Meshtastic Metrics Exporter reduced 500K+ Prometheus metric series to one Postgres database: consolidated eight queries into one with Tiger Data for 10,000 mesh radio nodes.
By Andrew Stebbins
September 16th, 2026
PostgreSQL write amplification explained: why indexes multiply WAL volume, how to measure it on your tables, and four ways to reduce it without schema changes.
By John Shi
September 11th, 2026
