Staff Platform Engineer - Developer Infrastructure

Persona AI Inc•Houston, TX

20h•Onsite

About The Position

Persona AI is building humanoid robots for demanding industrial environments, performing dangerous and physically demanding work. We are backed by leading investors and are engaged with global industrial leaders. Our work spans the robot platform and the systems required to deploy it at scale. As a Staff Platform Engineer, you will own the development infrastructure, which is currently supported by engineers who also write control code. Your success will be measured by improvements in build times, time-to-first-commit for new engineers, deployment frequency, and the reduction of infrastructure-related issues for the team. This role involves both cloud and physical infrastructure, including lab networks, robot development boxes, bench and HIL fixtures, on-site LAN infrastructure, and systems that support offline operation.

Requirements

8+ years of experience operating production infrastructure for a software engineering organization, with direct ownership of CI/CD or developer platform work.
Deep Linux systems fluency, including networking, storage, systemd, kernel, and driver debugging.
Hands-on ownership of a major CI system (e.g., GitHub Actions, GitLab CI, Buildkite, Jenkins) and understanding of the underlying build and cache layers.
Real-world experience with a large-scale C/C++ build system (e.g., Bazel, CMake, or equivalent), including cross-compilation and dependency pinning.
Proficiency in Infrastructure as Code (e.g., Ansible, Terraform) and a GitOps mindset for reviewable, reproducible, version-controlled changes.
Fluent in Python and Bash, with a strong ability to automate manual procedures.
Strong written communication skills, capable of documenting decisions that have long-term impact.

Nice To Haves

Experience shipping software to embedded or edge Linux targets (ARM64, Jetson, Yocto/custom images, A/B partitions, OTA update systems).
Experience with hybrid on-prem and cloud environments, with the ability to make clear judgments about resource allocation.
Familiarity with open-source observability stacks (Prometheus, Grafana, OpenTelemetry) and self-hosted services (registries, artifact stores, object storage).
Experience in robotics, autonomous vehicles, aerospace, or other hardware-intensive environments where release failures have physical consequences.
Knowledge of Nix, Bazel remote execution, or other tools for reproducible builds at scale.

Responsibilities

Build and CI: Develop the build graph for a mixed C++/Python/Rust/CUDA monorepo, focusing on incremental correctness, remote caching, and cross-compilation for AMD64 and ARM64 targets. Manage self-hosted CI runners, including GPU and hardware-attached runners, to ensure PR feedback remains under ten minutes. Ensure bit-identical artifact reproducibility from tagged commits.
Release and fleet delivery: Manage versioning, artifact promotion, and the container registry/package mirrors. Implement safe, resumable, and bandwidth-aware rollout processes for robots, including staged channels, canary robots, and reliable rollback over poor network links. Ensure signing and provenance for all software deployed to robots.
Developer platform: Create reproducible development environments across laptops, shared dev machines, and robots. Develop self-service tooling to enable autonomy engineers to deploy branches to robots without requiring tickets or deep Kubernetes knowledge. Streamline the onboarding process so new engineers can build, test in simulation, and deploy to a bench robot on their first day.
Infrastructure and observability: Manage cloud and on-prem compute, storage for multi-terabyte robot logs, and the operational layer of the training/simulation cluster. Implement fleet observability, including metrics, logs, and traces from robots to dashboards. Set up site infrastructure for deployments, such as VPN/overlay networking, local mirrors, and offline-capable registry authorization.