DevOps Engineer (local)

Pitch Aeronautics is a rapidly growing startup creating innovative solutions for the utility industry. They are seeking a DevOps Engineer to manage their AWS infrastructure, ensuring reliable software deployment and continuous delivery for their platform that processes real-time weather data.

Responsibilities:

• Build and manage CI/CD pipelines in GitHub Actions: build container images, push them to Amazon ECR, and promote them across environments using short-lived federated credentials, not long-lived keys
• Own our infrastructure as code: Terraform modules and remote state, nothing that lives only in the console, and a bill kept honest with budgets, billing alerts, and right-sizing
• Operate our hybrid compute footprint: Amazon ECS on AWS Fargate, plus Amazon EC2 instances running Dockerized workloads under systemd
• Own observability and centralized logging: our self-hosted Prometheus/Grafana stack, the log pipeline and search stack behind it, and alert rules worth acting on
• Operate our machine-learning services: deploy and run containerized inference for our customers, manage GPU instances, and keep multi-gigabyte model images and their registry lifecycle under control
• Own Amazon RDS for PostgreSQL: backups and point-in-time recovery, version upgrades, parameter groups, connection limits, and the slow-query conversation with the app team
• Own networking end to end: VPCs and segmentation, ALBs, NGINX in front of our Java apps, and the secure access paths engineers use to reach private infrastructure
• Implement and manage AWS security controls (IAM, encryption, and audit logging) and own the infrastructure side of our SOC 2 program: access control, patch automation, and vulnerability management

Requirements:

• 5+ years in DevOps, SRE, or platform engineering
• Design, deployment, and operation of scalable cloud architectures such as Amazon ECS in production (task definitions, service deploys, rollbacks), Amazon EC2, Amazon RDS, Amazon S3, VPC networking (private subnets, NAT vs. VPC endpoints, security groups, DNS), and Elastic Load Balancing
• Terraform in production, including writing modules, managing remote state, and handling drift
• GitHub Actions for building, testing, and deploying containerized services or equivalent experience you can transfer
• You've owned an alerting stack rather than just read its dashboards, and you can defend which alerts you deleted. Comfortable with Prometheus and Grafana, and with a log search stack (Elasticsearch/Kibana, OpenSearch, or similar)
• Python for automation, tooling, and diagnostics; Linux administration in production (Ubuntu, Amazon Linux)
• Least-privilege IAM design, secrets management, encryption, and supply-chain hygiene in CI
• Fluency with AI assistants in day-to-day engineering work, paired with a grasp of the concepts underneath them. Know what the models are doing and when and why they fail. You move faster with them, and you know enough to catch them when they're confidently wrong
• Keeping Maven-built Java/Spring applications healthy in production: heap and GC behavior
• Query and index tuning, RDS Performance Insights, and pg_stat_statements
• Serving models in production, managing the instances behind them, and orchestrating jobs (SkyPilot or similar)
• Familiarity with HRRR, GRIB, Zarr, or similar gridded datasets
• SOC 2, plus control automation, vulnerability scanning, and patch management at scale
• Supporting a product deployed into regulated customer environments which includes utility-sector security reviews, NERC CIP exposure, or third-party risk assessments