ContentRemoteRef: DL-2026-005
AI Engineer, Autonomous Content Systems
Build and operate the autonomous pipeline that researches, writes, illustrates, fact-checks and publishes our own technical content, end to end, with no human in the loop on a normal day.
What you'd do
- Own the agent pipeline that turns a sourced event (a security advisory, a model release, an incident writeup) into a published piece on devencelab.com: research, draft, illustrate, fact-check against citations, and publish, verified live, not just accepted.
- Encode house style as something an agent can actually be held to, not a document a human reads once: length and structure bounds, a banned-phrase list that blocks generated-sounding copy before it ships, a required pull quote, a required primary source.
- Build the guardrails that keep an autonomous publisher honest: a duplicate check against everything already published, a cadence limiter so a batch run cannot flood the site in one pass, a smoke test that publishes and then cleans up after itself without leaving test content live.
- Treat a generated article the way an engineer treats generated code: reviewed by a linter for tone and unsupported claims, checked for accessibility (every image carries real alt text or the build refuses to ship it), and traceable back to its sources.
- Extend the pipeline's own tool surface (currently around thirty MCP tools covering sourcing, drafting, illustration, SEO and verification) as new content formats or distribution channels come online.
What we're looking for
- Experience building agents or pipelines that produce real, checkable output autonomously, not demos that a human quietly cleans up before anyone sees them.
- Comfortable writing the evaluation and guardrail code around a generative system, not just the prompts: what makes a piece publishable, and how a machine checks that before it ships.
- Strong technical writing instinct yourself, enough to define and enforce a house style, spot when generated copy is technically correct but says nothing a reader didn't already know.
- Working knowledge of SEO and structured content basics (sitemaps, canonical URLs, machine-readable indexes) since this pipeline is also responsible for telling search engines and other agents what it published.
- Comfortable owning a system that runs unattended on a schedule: you will be the one who finds out why a 3am batch run behaved strangely, and fixes it before the next one.
Ready to apply?
Fill in the form on the next page, CV included; one submission sends it straight to us.