> Markdown version of [/jobs/ext/4509701-principal-software-engineer](https://www.wearedevelopers.com/jobs/ext/4509701-principal-software-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Principal Software Engineer

- **Company:** Mirantis, Inc.
- **Location:** Austin, TX, United States
- **Experience:** Expert
- **Contract:** Permanent contract
- **Skills:** Artificial Intelligence, Audit Trail, Continuous Integration, Software Debugging, Distributed Systems, Python (Programming Language), Autoscaling, Large Language Models, Kubernetes, Model Inference
- **Published:** November 1, 2026
- **Apply:** https://jobs.smartrecruiters.com/Mirantis/744000158796916-principal-software-engineer-

## About the Role

* 10+ years in infrastructure or distributed systems, including several as technical lead.
 * Shipped multi-service systems that enterprises run in production, including software they install themselves.
 * Expert Kubernetes: operators, CRDs, scheduling, Helm, and debugging at scale.
 * Production experience with GPU workloads, LLM inference or AI gateways
 * Expert Go, strong Python, and solid CI/CD and infrastructure as code.
 * Leads through writing and influence, uses AI-assisted development tools fluently, and thrives with autonomy on a distributed team.

## Description

As Principal Software Engineer, you are the senior technical voice for Enterprise AI. You own its architecture together with the teams that share the platform, and set the technical bar for a small, senior, remote-first team.

 * Own the architecture across the AI gateway, control plane, model serving, identity, metering and observability, for hosted and installed editions.
 * Lead the gateway strategy, including the data plane choice and policy-based routing with per-request metering and audit records.
 * Set direction for LLM serving on Kubernetes: GPU scheduling and sharing, autoscaling, multi-tenant and disaggregated serving, and cost per token.
 * Define how enterprises run the product: Helm installs and upgrades, air-gapped networks, preflight checks and supply-chain security.
 * Run design reviews, make build vs. buy calls, and record decisions as ADRs.
 * Build hands-on in the critical path, and mentor senior engineers, including on AI-assisted development.
 * Work with design partners and customers, and represent Mirantis in the Kubernetes and ML-infrastructure communities.