35+ YEARS IN MARKET
Continuously operating since 1990. 25,000+ consulting hours delivered and 596+ critical production incidents resolved. Pattern recognition only decades of real operation can build.
POSTGRESQL REMOTE DBA — 24×7
Dev team wrote the app, picked RDS or stood up Postgres on a VM and shipped. 6 months in: backup never tested, replica down, autovacuum off on a hot table. We take the operation.
PostgreSQL DBAs trusted by
POSTGRES WITHOUT AN OWNER
Backup that was never restored
RDS snapshot running daily. Nobody ever restored it to a parallel environment. When you really need it — logical corruption, accidental drop — that's when you find out if it works.
Streaming replica "up"
pg_stat_replication shows the replica connected. But replication lag hit 4 hours. Failover today means losing 4 hours of data. Nobody monitors wal_sender.
Autovacuum off to "not interfere"
Events table with 80M dead tuples. Transaction ID approaching wraparound. In 3 weeks the database stops by itself — and no alert was configured.
Major upgrade delayed 3 years
Still on Postgres 11. EOL passed. Every new query is tuned for 15+. Migration needs pg_upgrade or logical replication, with no long window.
TECHNICAL SCOPE
Not just CPU monitoring. Keeping Postgres healthy in production: real HA with Patroni, tested backup with pgBackRest, per-table autovacuum, on-call with Senior DBA — not first-line stumbling through a wiki.
Senior Postgres DBA on real standby. Sev1 SLA <15min, Sev2 <1h. No bot, no first-line, no "I'll open a ticket".
Patroni cluster setup and operation with etcd or Consul, automatic failover under 30s, planned switchover, DCS monitoring. HA that actually works when needed.
Incremental backup + WAL archive on S3/GCS, restore tested monthly in parallel, retention by compliance policy. Second-level PITR, proven.
Sync or async replica, real-time lag monitoring, sized replication slots, cascading replicas when sensible.
11→16 with in-place pg_upgrade or near-zero downtime logical replication. Regression tests, pre-upgrade plan capture, documented rollback.
Table growth analysis, I/O and RAM projection, RDS/Aurora or on-premise sizing. No upsell — only what fits the budget.
POSTGRES DBAs IN A PHYSICAL ROOM
Not a remote freelancer from an unknown country. DBA team with track record on Postgres since 8.4, operating in a controlled physical room at PIT/SP.
Backup isn't what you configured. It's the last restore you ran. If it was 6 months ago, you have no backup.
OPERATING IN PROD
BCB FINTECH
12-month uptime
3-node Patroni cluster operation on PostgreSQL 15 with pgBackRest, 12 months with no end-user-perceived incident.
HEALTHTECH SAAS
Restore proven
Point-in-Time recovery of accidental drop in production in 14 seconds via pgBackRest + WAL archive on S3. RPO window held.
SERIES B STARTUP
Zero-downtime upgrade
PostgreSQL 11 → 16 migration via logical replication with 45s switchover. Pre-plan capture + post monitoring.
METHODOLOGY
Environment audit
Instance, version, replica, backup, monitoring inventory. Prioritized risk map. We touch nothing — just read.
Quick wins
Backup with tested restore, critical alerts configured, autovacuum tuned on hot tables, replication lag monitored.
HA & DR
Documented Patroni or failover, pgBackRest with retention policy, incident runbook, active on-call.
Formal handover
Complete documentation, per-scenario runbooks, health dashboard, active SLA contract.
FAQ
All. Aurora PostgreSQL, RDS for PostgreSQL, Cloud SQL (GCP), Azure Database for PostgreSQL and on-premise. Tools differ (Patroni only outside managed), principles are the same.
We act as off-hours on-call, second opinion on critical decisions (upgrade, HA change), or surge during incident. We don't replace the team — we are technical backup.
You open an incident via WhatsApp, phone or email. Auto routing by severity. Sev1 (DB down): Senior DBA in 15min. Sev2 (degradation): 1h. Sev3 (question): next business day.
Monthly package by number of instances and SLA tier. No per-incident charge — wrong incentive. If volume changed, we talk before the month ends.
Yes. First 30 days are a formal handover with audit, quick wins and documentation. No rupture. If the previous DBA wants to participate, even better.
20 minutes with a Senior PostgreSQL DBA. Tell us the situation, we return a 30-day takeover plan. No sales form.
Versions and platforms supported
WHY HTI
Sustaining critical environments in Brazil since 1990. We're not a generic consultancy that learned databases — we are The Database Company.
Continuously operating since 1990. 25,000+ consulting hours delivered and 596+ critical production incidents resolved. Pattern recognition only decades of real operation can build.
The only MySQL Authorized Education Center (AMEC) of Oracle University in Brazil — achieved twice. Over 1,000 DBAs trained in official HTI courses.
DBAs operate from a restricted-access room at PIT São José dos Campos: 24h CCTV, biometric access, dedicated rack. No remote freelancers, no improvised workstations. Compliance that goes beyond internal policy.
Behavioral monitoring developed in-house. Detects anomalous ingestion patterns and access to protected objects — no third-party tool dependency.