Unlimit - Agentic Content Creator
Requirements
• A genuine writer's instinct. You can produce a clean, opinionated, useful piece of content from a blank page — and you can edit machine-generated drafts into something that actually reads well. Portfolio or samples expected. • Hands-on fluency with modern AI tooling. Daily user of Claude, ChatGPT, and at least one agentic or workflow tool (Claude Code, Cursor, n8n, Make, LangChain, or similar). You should be able to walk us through workflows you've built. • Light-to-moderate coding ability. Python or JavaScript at the level of "I can read API docs, write a script, debug it, and ship it." You don't need to be an engineer; you do need to not be blocked by one. • Prompt engineering as a real skill. You can explain why one prompt works and another doesn't, and you've built reusable prompt libraries or templates before. • Editorial judgement. You know when AI output is good, when it's mediocre-but-fixable, and when it needs to be thrown out and rewritten. You'd rather ship five great pieces than fifty average ones. • SEO/GEO literacy. You understand search intent, on-page structure, schema, internal linking, and what makes content citable. You don't need to be the strategist — but you need to speak the language. • Operator's mindset. Bias to ship. Comfortable owning a backlog, prioritising ruthlessly, and reporting on output without being managed. • Experience in our category ([fintech / payments / SaaS / your vertical]). • Familiarity with the Anthropic, OpenAI, or Google APIs for custom agent work. • Background in journalism, technical writing, or developer relations. • Open-source contributions, public prompt libraries, or writing about agentic workflows.
Responsibilities
• Design and operate agentic content pipelines. Build multi-step workflows — research → outline → draft → fact-check → structure → optimise → publish — using Claude, ChatGPT, Cursor, n8n / Make / Zapier, and custom scripts. You own the workflow, not just the output. • Produce GEO-shaped content at scale. Comparison pages, listicles, "best of" roundups, statistics hubs, glossaries, FAQs, definitive guides, and structured Q&A — formatted for both human readers and LLM retrieval. Volume matters; so does quality. Both, not either. • Run off-site content campaigns. Draft and orchestrate the long tail of third-party touchpoints that build LLM-trusted authority — Reddit threads, Quora answers, niche forums, GitHub READMEs, dataset descriptions, review platforms. Always ethically, always disclosed where required. • Build internal tools and agents. Prototype the small utilities that save the team hours: prompt-tracking scrapers, citation monitors, content gap analysers, brand-voice linters, fact-checking agents. Think "useful internal tool" not "production SaaS." • Maintain editorial quality and brand voice. Agentic doesn't mean autopilot. You're the final editor on everything that ships. You design the QA loop — automated checks plus your own eye — that keeps the bar high. • Fact-check rigorously. LLM-assisted content fails the moment it hallucinates a stat or misattributes a source. You build verification into the workflow and personally vet anything sensitive. • Iterate on prompts and templates as a craft. Treat prompts and agent instructions like code: versioned, tested, documented, refactored. Share what works with the rest of the team. • Partner with the GEO Specialist. They'll bring the strategy and the target prompts. You translate that into a steady output of content that earns citations and visibility.
Apply in one click
Upload My Resume
Drop here or click to browse · Tap to choose · PDF, DOCX, DOC, RTF, TXT