This is the section where round two of the screen actually happens, the closing gate before an
interview hits your inbox. A recruiter takes their time here, and even at that, the current
role still drives around 95% of the result.
That tracks: nothing proves what you can run in production today like the seat you sit in
right now. To earn a "yes", the section has to hit every entry on the
Database Admin role profile, one bullet per domain you named in Domain
Expertise above. Every bullet has to come off something you genuinely held in production,
never a ticket that landed on your queue.
1
Database Administration & Schema Design
The flagship work of the role. Show the engines you operate, the version mix you keep
current, the schemas you steward, and the change-review process behind any new index,
table, or constraint. Name the engine and the schema you own, not "administered
databases".
Techniques
Schema design & review
Index strategy
Partitioning
User & role management
Tools
PostgreSQL 14 / 15 / 16
MySQL 8, MariaDB
SQL Server 2019 / 2022
Metrics
Databases under management
Data volume
Schema-change throughput
2
Query Performance & Tuning
Where Database Admin time actually goes. Show the slow-query review you run, the
offending plans you fixed, the index you added or dropped, and the application work you
steered to a better access pattern. Name the query class and the p99 you cut, not
"tuned queries".
Techniques
EXPLAIN plan analysis
Index design
Statistics & vacuum
Query rewriting
Tools
pganalyze, pg_stat_statements
Percona PMM, slow query log
Datadog DBM, SolarWinds DPA
Metrics
p95 / p99 latency cut
CPU per query reduced
Slow-query count down
3
Replication, HA & Failover
How the database survives a node loss without paging every engineer. Show the topology
you maintain, the replica lag you keep in check, the failover drills you run, and the
real switchover you led. Name the topology and the failover time, not "managed
replication".
Techniques
Streaming replication
Automated failover
Read-replica routing
Switchover drills
Tools
Patroni, repmgr, pg_auto_failover
MySQL Group Replication, Galera, Orchestrator
SQL Server Always On, MSSQL AGs
Metrics
Availability held (four 9s)
Failover time cut
Replica lag p99
4
Backup, Recovery & Disaster Recovery
What turns an "we're going to lose the company" outage into a 20-minute
restore. Show the backup platform you run, the point-in-time recovery you prove every
quarter, and a real restore you led under pressure. Name the recovery objective and the
actual restore, not "managed backups".
Techniques
RPO / RTO design
Point-in-time recovery
Cross-region snapshots
Restore drills
Tools
Barman, WAL-G, pgBackRest
MySQL Xtrabackup, mysqldump
SQL Server backups, RMAN
Metrics
RPO / RTO held
Restore success rate
Data volume under restore testing
5
Capacity Planning & Storage
How the database keeps growing without falling over. Show the storage you sized, the
disk-IO profile you defend, the connection-pool tuning behind it, and the resize you
led without downtime. Name the capacity event you handled, not "monitored
storage".
Techniques
Storage sizing
Connection pooling
Tablespace planning
Workload forecasting
Tools
pgBouncer, ProxySQL, HAProxy
EBS, gp3 / io2, Aurora storage
NetApp, Pure, Dell PowerStore
Metrics
IOPS headroom
Connection ceiling held
Storage cost per GB cut
6
Database Security & Compliance
How the data stays accountable. Show the role-based access you enforce, the encryption
(at-rest and in-transit) you maintain, the audit logging your auditors actually read,
and the compliance program (SOC 2, PCI, HIPAA) you carry. Name the control and the
audit it cleared, not "handled security".
Techniques
Role-based access
TDE & TLS in transit
Audit logging
Data masking
Tools
Vault, AWS KMS, CMKs
pgaudit, MySQL Enterprise Audit, SQL Server Audit
Imperva, DataSunrise
Metrics
Audits passed
Privileged-access reviews held
Sensitive-field coverage
7
Migration & Upgrade Programs
The big swing-the-pickaxe work: a major-version upgrade, an on-prem to cloud move, an
engine swap, or a sharding rollout. Show the program, the online-cutover technique you
used, and the downtime budget you stayed under. Name the migration and the downtime,
not "led an upgrade".
Techniques
Online schema change
Logical replication cutover
Blue-green database
Engine swap
Tools
pg_upgrade, pg_repack
gh-ost, pt-online-schema-change
AWS DMS, Cloud SQL DMS, Azure DMS
Metrics
Downtime budget held
Data migrated
Rollback rate
8
Monitoring, Tooling & Automation
What turns a 4-person team into one that scales with the workload. Show the
schema-as-code workflow, the migration tooling in CI, the dashboards on-call actually
opens, and the routine chore (vacuum, reindex, replica rebuild) you put into a
playbook. Name the artifact and the toil reclaimed, not "automated tasks".
Techniques
Schema as code
Migration CI
Vacuum & reindex jobs
Runbook libraries
Tools
Liquibase, Flyway, Schemahero
Ansible, Terraform DB providers
Prometheus exporters, Grafana, PagerDuty
Metrics
Toil hours reclaimed
Migrations per week
Mean time to acknowledge