Logo for Hydra Host

Database Platform Engineer

Role overview

Summary

This role focuses on database management within Hydra Host's Platform Team, ensuring the integrity and security of applications across the company's infrastructure. You will develop and implement patterns for database provisioning and management, serving internal developer teams and enhancing operational efficiency.

Qualifications

  • 5+ years running Postgres or MySQL in production.
  • Experience automating database lifecycle with Terraform.
  • Strong HashiCorp Vault experience.
  • Experience with managed database platforms and self-hosted databases.
  • Ability to read HCL, Python, TypeScript, Go, shell, and SQL.

Responsibilities

  • Develop patterns for securing, consolidating, optimizing, and scaling databases.
  • Own the canonical patterns for database provisioning, credentials, backups, restores, migrations, and reversions.
  • Manage and consolidate database modules in the Infrastructure as Code catalog.
  • Implement change management processes for database migrations and ensure compatibility.

Key facts

Hard skills

Other skills

  • Problem Solving

About the company

Hydra Host logo

Hydra Host

  • Cloud & IT Infrastructure
  • Startup
  • 11–50
Hydra Host specializes in providing data centers with unparalleled provisioning, management, and security solutions, allowing them to focus on their core competency - hardware and infrastructure. By partnering with Hydra Host, data centers will enhance their customer experience, streamline operations, and stay ahead of their competition with our innovative products and services. Reach out to sales@hydrahost.com to learn how Hydra Host can take your business to the next level!

Your match analysis

See how your profile stacks up against this role.

We compared the job requirements to your profile to show where you're strong and where you fall short.

Job description

About Hydra Host

Our Company

Hydra Host is a Founders Fund-backed NVIDIA Cloud Partner building the infrastructure platform that powers AI at scale. We connect AI Factories, high-performance GPU data centers, with the teams that depend on them: research labs training foundation models, enterprises running production inference, and developer platforms that need compute capacity on demand. 

The Role

Hydra Host's Platform Team owns our internal developer portal: change management, full-stack release candidates, self-service rollbacks, and disaster recovery. Database management is where those promises go to die. A rollback that ignores the schema is a lie; a DR plan that ignores the data is a wish. We are hiring the engineer who makes every lifecycle promise hold for every database we run. As the company scales, database management becomes central to the integrity and security of every application across our footprint.

This is a pattern-building role. Every database already deploys through CI/CD with backups and migrations, integrated with our automated secrets and identity engines. You will develop those patterns to secure, consolidate, optimize, and scale each database we run. You will own the canonical patterns for how dev teams provision databases, credentials, backups, restores, migrations, and reversions, and you will build the self-service features in our IDP that dev teams use to manage them. You will also be the subject matter expert on all things database: current on business needs and industry standards, with a clear opinion on every architecture question.

What You'll Own

One pattern per engine

  • Dev Patterns: Postgres, MySQL, Redis, Clickhouse, and the next engine someone adopts: each gets a single shared module covering provisioning, config, monitoring, backups, and teardown protection. A dev team invokes from our IDP, not a recipe a platform engineer follows.
  • Database Modules in IaC Catalog: manage and consolidate the ones we have; build more as needed.
  • Consolidation: you own the move to consolidate databases to one server per engine and the controls that make sharing safe: per-service roles and grants in code, dynamic credentials per service, network allowlists, per-tenant restore, and an audit trail per database.
  • Cost Scaling: database spend should scale with load, not with the service count.
  • DB hosting architecture: you own the decision where a database should be hosted: managed service or self-hosted, how and why.
  • Protection of database content: survival beyond deletion events; rollbacks; access; and everything else related to reliability and security of database content.
  • Change Management: forward-only, backward-compatible migrations enforced in the shared pipeline: migration tests on scratch databases, a schema diff on every merge request, and unsafe DDL gated behind an explicit expand/contract acknowledgment.
  • Backup and Recovery: the backup jobs and the restore tests that prove them, with point-in-time restore and forking as self-service actions in our IDP.
  • Access: short-lived, per-service credentials from Vault in place of static passwords; human admin access as a short-TTL escalation through our IDP and Zitadel; roles and grants in code; an audit trail that answers who touched which database, when, with what credential.

Requirements

  • Self-starting attitude and extreme ownership — you see a gap in the platform, you close it
  • 5+ years running Postgres or MySQL in production: replication, failover, backup and restore, connection pooling, and enough scar tissue to distrust a green health check
  • You automate database lifecycle with Terraform and can tell from a plan when it is about to eat data
  • You have designed forward-only, backward-compatible migration flows (Drizzle, Prisma, Alembic, Atlas, Flyway, or similar) and can explain why "just run the down migration" is not a rollback strategy
  • You have run both managed database platforms (Aiven, RDS, Cloud SQL, Redis Cloud) and self-hosted databases, and can defend a choice between them for a given workload
  • Strong HashiCorp Vault experience: the database secrets engine, policies, auth methods, lease lifecycle
  • You have built OIDC-backed access to infrastructure with Zitadel, Keycloak, Azure AD, Okta, or similar
  • Stateful workloads on an orchestrator (Nomad or Kubernetes) do not scare you
  • You read HCL, Python, TypeScript, Go, shell, and SQL without complaint — all are in rotation here
  • You have run database automation in production and know where it breaks: migrations through a connection pooler, instances racing the same migration, lock contention, credential rotation that drops the pool, provider quirks that turn a rename into a destroy. You design so a failed run never leaves a database half-migrated, credential-less, or unrecoverable.

Preferred

  • Nomad and nomad-pack, not just Kubernetes
  • Zitadel, or another self-hosted OIDC provider you ran yourself
  • PgBouncer or another pooler sitting in front of automated credential rotation
  • Aiven: its Terraform provider, forking, and PITR.
  • Restore-tested backup tooling: restic, pgBackRest, WAL-G
  • Data infrastructure beyond OLTP: ClickHouse, Kafka, CDC/Debezium
  • You have produced compliance evidence for database access or change management (SOC 2 or similar)

Salary: 150000 - 200000 USD per year

Related jobs

See all Platform Engineer jobs

Other jobs at Hydra Host