Forward Deployed Engineer, Infrastructure and Deployment

Job description

# **About Preql** Preql helps enterprises clean, unify, and govern messy internal data so it actually works for AI, analytics, and reporting. We work with large organizations navigating complex data environments and high-stakes operational workflows. Based in New York, our team comes from data infrastructure, AI, and enterprise software. ## **How we work** We’re a small team with little bureaucracy. Leadership expects individuals to take ownership, move quickly, and make good decisions for the company with support from their teammates. The curious do well here, are comfortable operating in ambiguity, and are willing to form opinions and act on their convictions instead of waiting for instructions. ## The role We ship self-hosted software into regulated enterprises. That means every deployment involves someone else's Kubernetes cluster, someone else's identity provider, someone else's network policy, and a security team that has to sign off before any of it runs. You will own how Preql gets installed, secured, upgraded, and kept healthy in customer environments. **What you will own** - Installation and configuration of Preql in customer environments: our Docker images and Helm charts, deployment into customer managed Kubernetes across AWS, Azure, GCP, and on-premise - Networking, identity, and access setup: private connectivity, SSO and SAML, IAM and role design, warehouse permissions, secrets management - The deployment architecture for each account, including choosing the right configuration and making sure what is scoped in the SOW is what actually gets built - Navigating customer IT environments to proactively uncover and resolve potential blockers - Enterprise security and compliance review: questionnaires, data handling and residency requirements, architecture walkthroughs with customer security teams, and the escalations that come with regulated buyers - Production health in customer environments: monitoring, upgrades, and first response when something breaks, rather than escalating straight to product engineering - Release and versioning discipline that keeps every customer on a known, supportable configuration - Runbooks, install automation, and reference architecture documentation that make each deployment faster than the last ## What success looks like - **90 days:** you have run an install end to end without product engineering in the room, and you can walk a customer's security team through our architecture yourself - **6 months:** install time for a comparable customer has dropped measurably, every account is on a known version, and there is a runbook that did not exist before - **12 months:** deployment is a repeatable process rather than a project, and product engineers are not being pulled into customer environments ## What we are looking for - 5+ years in infrastructure, platform, or DevOps engineering, shipping into production environments you did not control - Docker and Kubernetes in production - Depth in at least one of AWS, Azure, or GCP, and a working understanding of the constraints in the others - Networking and identity in enterprise settings: VPCs and private connectivity, SSO and SAML, IAM and role design, secrets management - CI/CD, observability, and incident response - High SQL and data infrastructure literacy - Comfort working directly with customers, including scoping, pushing back, and delivering bad news early - High tolerance for ambiguity. Early deployments will not have a runbook, and you will write the runbook ## Strong signals - You have packaged and shipped self-hosted or customer managed software, not just SaaS - You have carried a deployment through a bank or other regulated buyer's security review and kept the project moving while it was in flight - Familiarity with enterprise data infrastructure and the finance systems around it