Systems and Automation Engineer

Role overview This position focuses on owning the automation backbone that powers both client engagements and internal systems. The work centers on replacing fragile, one-off scripts with durable, observable workflows that are designed to be retried, audited, and handed off cleanly. It is a hands-on systems and platform role for someone who treats automation as a product rather than a side effect.

Responsibilities - Design and operate scheduled and event-driven workflows across runtimes such as GitHub Actions, Cloud Run jobs, AWS Lambda, and durable workflow engines, with retries and idempotency as first-class concerns. - Build typed integration scripts in TypeScript or Python that connect commerce platforms, content management systems, data warehouses, and messaging tools, with explicit contracts at every boundary. - Add LLM-driven steps where they earn their place (classification, extraction, summarization, content generation), paired with schema validation and human review gates. - Instrument every job with structured logs, run history, and actionable alerts that name the failing step and likely fix, and build guardrails such as dry-run modes, staged writes, rate limits, and kill switches. - Manage secrets, credential rotation, and least-privilege service accounts for everything that runs unattended, and own the operational hygiene that keeps unattended work safe. - Consolidate scattered one-off scripts into shared, documented tooling, write runbooks so any engineer can rerun, backfill, or pause a pipeline, and audit what is running, retire what is unused, and keep automation costs honest.

Requirements - Four or more years in systems, platform, DevOps, or backend engineering with genuine production-automation ownership. - Strong scripting in TypeScript or Python, working knowledge of shell, and the judgment to recognize when a script should be promoted to a service. - Hands-on experience with GitHub Actions and at least one cloud job runtime such as Cloud Run, Lambda, ECS, or Kubernetes CronJobs. - Comfort with Postgres, message queues, and the failure modes of third-party APIs, including the discipline to fix logging after a 2 a.m. incident so the next on-call inherits a calmer system.

Nice to have - Experience using LLM APIs in batch or pipeline contexts, including cost control and output validation. - Familiarity with workflow engines such as Temporal, Inngest, Airflow, or Prefect. - Exposure to commerce or content operations, for example Shopify, Sanity, or another headless CMS.