About Mariana Minerals

Mariana Minerals is a software-first, vertically integrated minerals company on a mission to supply the critical minerals powering modern energy, AI, and defense technologies. We’re reimagining the minerals supply chain by combining deep industry expertise with advanced software, automation, and data-driven decision-making.

The role

Mariana Minerals is looking for a Sr. or Staff Software Engineer, Infrastructure to join the team that owns and collaborates on the platform every other engineer here builds on.

Our software team builds MarianaOS, the operating system for how Mariana engineers, constructs, and operates its mines and mineral processing facilities. It runs in AWS and on edge clusters at our sites, including our copper site in Moab, Utah and our lithium facility in East Texas. On any given day, it serves construction dashboards, running GPU inference on site camera feeds, training models for a lithium extraction circuit, and moving plant floor data into a warehouse.

Your team owns the platform underneath all of that: clusters, deployment, CI/CD, observability, site/edge footprint, identity, secrets, and cloud costs. You will also collaborate with machine learning engineers and data engineers on the ML compute platform and data pipeline infrastructure.

The part of this role we care most about is what you build for other people. We do not want an approval queue. When an engineer needs a database, a bucket, a dashboard, or an alert, they should get it from a documented path they can run themselves, and your job is to make that path exist and make it the easiest option. You will be measured on how much infrastructure work happens without you.

We move quickly. Our engineers ship, get feedback from the people running our plants and job sites, and iterate. Everything you build should shorten that loop.

Our stack is TypeScript and React on the front end, Node and Python services, Postgres, and AWS. The platform is EKS with Karpenter, ArgoCD, Helm, Pulumi, Prometheus and Grafana, Loki, Prefect, Snowflake, and a growing amount of LLM and agent tooling in production.

What You'll Do

Own our Kubernetes footprint end to end: the EKS cluster, the on-premises site clusters, cluster upgrades, capacity and autoscaling, ingress and DNS, and the GitOps delivery path that turns a merged commit into a running service.

Build self-service infrastructure for other engineers, so that provisioning a service, a database, a preview environment, or a scheduled job is a documented path they run themselves so the team can deliver quickly.

Make observability a default. Every deployed service should emit metrics, ship logs, trace requests, and carry meaningful alerts without its author's writing bespoke plumbing.

Extend the platform to our sites, running on-premises and edge clusters that host GPU inference and plant data collection over constrained links.