Platform Engineer - AI/ML Infrastructure (Kubernetes & Terraform)
Deepgram · Remote · USA - Remote ·
COMPENSATION
$190k – $240k
Listed on the employer's job board (Ashby)
- Fully remote
- Kubernetes & Terraform
- AWS + Bare Metal
- AI/ML Infrastructure
- Senior Level
Join Deepgram to architect and operate the foundation for real-time Voice AI. You will build a self-service platform spanning AWS and on-premise data centers, enabling researchers to train and deploy complex models at scale. The team focuses on high-performance GPU orchestration and robust Infrastructure-as-Code practices.
What you'll do
- Architect core computing platforms using Kubernetes on AWS and on-premise
- Manage entire infrastructure via Terraform for reproducibility and automation
- Optimize AI/ML job scheduling by integrating Slurm with Kubernetes clusters
- Provision and maintain bare metal servers for high-performance GPU computing
- Implement networking and storage solutions for low-latency hybrid workloads
- Build observability stacks for monitoring, logging, and incident response
- Collaborate with researchers to create tools that accelerate development cycles
- Automate the lifecycle of single-tenant managed deployments
What you'll need
- 5+ years in Platform Engineering, DevOps, or Site Reliability Engineering
- Hands-on experience building production infrastructure with Terraform
- Expert knowledge of Kubernetes architecture in large-scale environments
- Strong scripting skills in Python, Go, or Bash
- Experience with CI/CD systems like GitLab CI, Jenkins, or ArgoCD
Nice to have
- Experience with Slurm for managing GPU-intensive AI workloads
- Background in bare metal provisioning using PXE boot or MAAS
- Familiarity with FinOps and cloud cost optimization strategies
- Knowledge of Kubernetes networking (Calico, Cilium) and storage (Ceph)
- Experience working in multi-region or hybrid cloud environments
Stack
- Kubernetes
- Terraform
- AWS
- Slurm
- Python
- Go
- Bash
- GitLab CI
- Jenkins
- ArgoCD
How you'll work
- Fully remote position
- Expects active use of AI tools in daily workflows
- Fast-paced environment requiring constant adaptation