Prompt Engineer (Marketing) Job Description

A starting template. Tailor to your stack, content surface area, and AI maturity.

What this role exists to do

You'll build the prompt library the marketing team ships on. You'll design, evaluate, and version prompts for content, lifecycle, performance, brand, and operations use cases. You'll turn what works into reusable templates the rest of the team can pick up without your help.

The role is engineering-minded content craft. Depending on team structure, you'll partner with the AI Marketing Manager, AI Content Strategist, or both to translate fuzzy marketing goals into structured prompts, evaluation harnesses, and small internal tools. You'll run experiments against models from providers like Anthropic, OpenAI, and Google, and you'll ship the prompt and tooling artifacts the team relies on every day.

What you'll be doing

• Designing and shipping prompts and prompt templates for marketing use cases across the funnel: long-form content, lifecycle email, ad copy, paid social, landing pages, brief generation, research synthesis.
• Building and maintaining the central prompt library. Versioned, documented, discoverable.
• Defining evaluation harnesses for marketing prompts: rubrics, golden examples, automated checks, human review loops.
• Running rigorous A/B tests across models, prompt variants, and parameter settings. Documenting what worked, what didn't, and why.
• Building retrieval-augmented generation (RAG) workflows over brand and content knowledge bases when prompting alone isn't enough.
• Prototyping small internal tools that wrap prompts (briefing assistants, voice-checkers, research summarizers) using model APIs and frameworks like LangChain where appropriate.
• Partnering with the AI Content Strategist on voice fidelity, with the AI Marketing Manager on roadmap, and with engineering on production deployment and observability.
• Training marketers on prompt patterns. Writing playbooks. Running office hours.
• Staying close to the model and tooling landscape (including open models on Hugging Face) and bringing options to the team with clear tradeoffs.

What you'll bring

The non-negotiables:

• 3+ years of hands-on work with generative AI, including at least one year shipping prompt systems in production.
• A portfolio of prompts and prompt-driven tools you've shipped, with documented evaluation methods and outcomes.
• Strong written communication and an eye for voice. You can rewrite a flat AI draft into something on-brand without prompting tricks.
• Working knowledge of model APIs, basic Python or JavaScript, and the lifecycle of a prompt from prototype to production.
• Comfort with evaluation: rubrics, A/B testing, golden sets, and basic statistical reasoning.
• Familiarity with at least one RAG implementation pattern and the tradeoffs involved (chunking, embeddings, retrieval quality, grounding).
• Strong cross-functional collaboration. You translate between marketers, designers, and engineers.

What helps but isn't required:

• Background in marketing, content, or copywriting before pivoting to AI work.
• Experience fine-tuning or building agents on top of model APIs.
• Familiarity with LLM observability and cost optimization at scale.
• Experience with structured output, tool use, and function calling.
• Background in regulated industries where prompt safety and accuracy matter.

What success looks like

By day 30. You've audited existing prompts, tools, and AI workflows. You've talked to content, lifecycle, performance, and brand leads. You know where the highest-leverage gaps in the prompt library are.

By day 60. Central prompt library v1 is shipped, with versioning and documentation. At least one evaluation harness is in place with a real golden set. One production-grade prompt-driven tool is live.

By day 90. Training and office hours are running. The first quarterly report on prompt performance, tooling adoption, and impact metrics is published.

What we offer

[Customize: comp range, equity, remote/hybrid policy, benefits, model API and tooling budget the role will own.]

How to apply

[Customize: portfolio expectations, take-home overview, hiring contact.]

Looking for a Prompt Engineer with marketing fluency? Hire Digital places vetted AI-native talent. Companion docs: Hiring Guide · Interview Questions.