FDA - Platform Engineer

We need to output HTML fragment with sections for headings: About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule. Only include sections that have real specific content. Must not repeat job title. Must not include Job ID etc. Remove boilerplate. Must preserve all substantive info. We need to parse the posting. Let's extract sections: - About the role: The paragraph after "Make An Impact On Federal Health Join Rackner to build and operate the secure, cloud-native platform..." That's description. So we can put that in

. - Responsibilities: The bullet list under "What You'll Do". Those are responsibilities. - Requirements: The bullet list under "Who You Are". Those are requirements/qualifications. - Qualifications: Might be overlapping with requirements; but we can treat "Who You Are" as Requirements. Possibly also "Preferred / Bonus" as Qualifications or Skills? Actually "Preferred / Bonus" are bonus qualifications. Could be Skills. - Skills: Could list specific technologies: Kubernetes, Terraform, ArgoCD, etc. But we already have them in requirements. Might combine. - Benefits: The section "Why You'll Love Working Here" includes Weekly Pay & Fully Remote, Professional Growth, Comprehensive Benefits, Work-Life Perks. Also "Weekly Pay & Fully Remote" indicates pay schedule? It says Weekly Pay & Fully Remote. So Pay: Weekly. Schedule: Fully Remote (maybe schedule? Actually schedule is remote). But we need separate Pay and Schedule sections only if posting actually states them. It states Weekly Pay & Fully Remote. So we can have Pay: Weekly. Schedule: Fully Remote (or Remote). However schedule might be remote work arrangement. We'll include. - Benefits: Comprehensive Benefits – 401k (100% match up to 6%), PTO, medical/dental/vision, life & disability insurance; Work-Life Perks – Gym/fitness membership, home office setup, swag, snacks, and social events. Also Professional Growth – Paid certifications and training. Thus we can create Benefits section with those details. We must not include boilerplate like hashtags, etc. We must not repeat job title. We must not include "Clearance: U.S. Citizenship required; must be able to obtain and maintain a Public Trust (active Public Trust a plus)" - that's a requirement; include in Requirements. We must not include "Location: Remote (U.S.)" - that's location; maybe include in schedule? Actually schedule is remote. Could include location in a short paragraph? The instruction: do NOT put Job ID / requisition numbers as headings — fold minor metadata into a short

or drop it. Location is minor metadata; we could include in a short

maybe under About the role? But we already have About the role description. Could add a short p: "Location: Remote (U.S.). Clearance: U.S. Citizenship required; must be able to obtain and maintain a Public Trust (active Public Trust a plus)." That seems okay. But we must not invent or summarize away detail. Must preserve all substantive information. So we need to include that clearance requirement somewhere. Could be in Requirements as a bullet. Let's structure:

About the role

Make An Impact On Federal Health Join Rackner to build and operate the secure, cloud-native platform that powers data, analytics, and AI/ML for a federal health mission. You won't just run infrastructure — you'll own the internal developer platform that data scientists, data engineers, and application teams build on every day. Shape where it goes next: production-grade MLOps, self-service onboarding, and a paved road from commit to GovCloud.

Location: Remote (U.S.). Clearance: U.S. Citizenship required; must be able to obtain and maintain a Public Trust (active Public Trust a plus).

Responsibilities

• Own the IaC + GitOps foundation — evolve Terraform and ArgoCD (ApplicationSets) so the platform is declarative, self-healing, and recoverable from scratch
• Build and operate a multi-tenant, self-service platform — golden-path app onboarding, tenant/environment management, an app & tooling catalog, and internal developer tooling/CLI
• Run Kubernetes on AWS GovCloud — Karpenter and KEDA autoscaling, ingress and Gateway API, cert-manager, and ALB integration across dev/pre-prod/production
• Architect and deploy AIOps/MLOps for data science and data engineering teams — GPU compute and scheduling, model training and serving, managed notebooks (e.g., SageMaker), GenAI/Bedrock workloads, and production-grade ML pipelines
• Enable the data lakehouse — Trino/Iceberg on Glue + S3, Dagster orchestration, dbt, and search — so teams have reliable, performant data access
• Deliver observability and SLOs — Prometheus, Grafana (including custom-built dashboards), and actionable alerting
• Automate security and compliance — secrets management, network policy, IRSA least-privilege, policy-as-code, vulnerability remediation, and ATO artifact generation
• Drive documentation and knowledge transfer — author and maintain platform docs, production-readiness standards, and runbooks
• Partner directly with government stakeholders and platform tenants — translate mission needs into platform capabilities, with room to grow into a technical leadership role

Requirements

• 7+ years in Platform Engineering, DevSecOps, Site Reliability, or Cloud Engineering, with senior-level ownership of a production platform
• Built an internal developer platform / PaaS from the ground up for federal clients — and taken it to production (through ATO) in a federal program, not just contributed to one
• Extensive hands-on experience across both the open-source cloud-native/data ecosystem (Kubernetes, Terraform, ArgoCD, Karpenter/KEDA, Prometheus/Grafana/Loki, Trino, Dagster, dbt) and the AWS services that back them (EKS, S3, Glue, IAM/IRSA, ECR, SageMaker, Bedrock, Athena)
• Deep expertise with Kubernetes (EKS), GitOps (ArgoCD or Flux), and Terraform infrastructure-as-code
• Designs for self-service: golden paths, multi-tenancy, and onboarding that lets teams ship safely without a platform engineer in the loop
• Strong CI/CD background (GitLab CI, Jenkins, ArgoCD, and progressive delivery)
• Fluent automation and scripting (Python, Bash, Go)
• Working knowledge of federal compliance (NIST 800-53, FISMA, ATO processes); comfortable in GovCloud/regulated environments
• Treats documentation and knowledge transfer as part of "done" — clear docs and runbooks come naturally
• A people person and problem-solver who's comfortable being client-facing, wearing many hats, mentoring teammates, and growing into technical leadership

Preferred / Bonus

• Hands-on MLOps/AIOps production experience for DS/DE teams — SageMaker, Bedrock, Kubeflow, MLflow, Ray, KServe/Triton/Seldon, or GPU workload scheduling
• Experience with federal health or other regulated federal data programs
• Data platform / lakehouse experience — Trino/Presto, Iceberg, Spark, Dagster/Airflow, dbt
• AWS GovCloud, IL4/IL5, FedRAMP, or air-gapped environment experience
• Policy-as-code (OPA/Gatekeeper, Kyverno), container security, SBOMs, and vulnerability scanning
• Kubecost/cloud cost management
• Experience establishing a documentation practice and production-readiness standards

We could treat Preferred / Bonus as Skills or Qualifications. But we already have Requirements. The instruction: Use headings for section headings (e.g. About the role, Responsibilities, Requirements, Qualifications, Skills, Benefits, Pay, Schedule). So we need to map content to those headings. We have About the role, Responsibilities, Requirements. We could put Preferred / Bonus under Skills or Qualifications. Let's decide: Qualifications could be the preferred/bonus. Skills could be specific technologies list. But we already listed many tech in Requirements. Might be redundant. Better: Use Requirements for mandatory, Qualifications for preferred/bonus, Skills for specific tech list (maybe summarizing). But we must not create thin sections. Let's see if we have enough content for Skills: we could list specific technologies: Kubernetes (EKS), Terraform, ArgoCD, Karpenter, KEDA, Prometheus, Grafana, Loki, Trino, Dagster, dbt, AWS services (EKS, S3, Glue, IAM/IRSA, ECR, SageMaker, Bedrock, Athena), etc. That's substantive. But we already listed them in Requirements bullet. Could move them to Skills. However we must preserve all info; moving is okay as long as not lost. We need to avoid duplication. Let's think: The posting includes both mandatory experience and specific tech. We could keep Requirements as the experience/years etc., and Skills as the tech stack. Qualifications could be preferred/bonus. Thus:

Requirements

• 7+ years in Platform Engineering, DevSecOps, Site Reliability, or Cloud Engineering, with senior-level ownership of a production platform
• Built an internal developer platform / PaaS from the ground up for federal clients — and taken it to production (through ATO) in a federal program, not just contributed to one
• Designs for self-service: golden paths, multi-tenancy, and onboarding that lets teams ship safely without a platform engineer in the loop
• Strong CI/CD background (GitLab CI, Jenkins, ArgoCD, and progressive delivery)
• Fluent automation and scripting (Python, Bash, Go)
• Working knowledge of federal compliance (NIST 800-53, FISMA, ATO processes); comfortable in GovCloud/regulated environments
• Treats documentation and knowledge transfer as part of "done" — clear docs and runbooks come naturally
• A people person and problem-solver who's comfortable being client-facing, wearing many hats, mentoring teammates, and growing into technical leadership

Skills

• Deep expertise with Kubernetes (EKS), GitOps (ArgoCD or Flux), and Terraform infrastructure-as-code
• Extensive hands-on experience across the open-source cloud-native/data ecosystem: Kubernetes, Terraform, ArgoCD, Karpenter/KEDA, Prometheus/Grafana/Loki, Trino, Dagster, dbt
• Extensive hands-on experience with AWS services that back them: EKS, S3, Glue, IAM/IRSA, ECR, SageMaker, Bedrock, Athena
• Experience with GPU compute and scheduling, model training and serving, managed notebooks (SageMaker), GenAI/Bedrock workloads, production-grade ML pipelines
• Experience enabling data lakehouse: Trino/Iceberg on Glue + S3, Dagster orchestration, dbt, search
• Observability: Prometheus, Grafana (custom dashboards), alerting
• Security/compliance automation: secrets management, network policy, IRSA least-privilege, policy-as-code, vulnerability remediation, ATO artifact generation

Qualifications