This role is about designing the core AI system that powers copy generation. It is not an infra-only role, not a prompt-only role, and not about single-line captions. The work involves translating persuasion and human judgment into scalable AI systems.
The position starts as fractional/consulting with a clear path to a full-time leadership or CTO role for the right person.
## Responsibilities - Design and build AI systems for multi-frame social media copy - Implement pipelines, routing, constraints, and evaluation layers - Work directly with LLM APIs (OpenAI, Anthropic, etc.) - Diagnose and fix why AI outputs fail at a system level - Define quality standards for shippable outputs - Collaborate with product and strategy as the system evolves
## Who we’re looking for - Experience building AI systems for multi-frame social media content - Ability to preserve hooks, pacing, and narrative flow across sequential posts - Can improve output quality through system design, not just prompts - Able to build production-level systems - Understand persuasion, copywriting, and human judgment in addition to LLM mechanics - Take ownership of outcomes and quality
## Not a fit if - AI experience is limited to single captions, blogs, or generic content - You focus only on backend infrastructure or DevOps - You mainly identify as a prompt engineer - You need very detailed specs before starting work - You are only interested in short-term freelance projects
## Technical requirements - Production software experience - Hands-on with LLM APIs - Experience building internal tools, products, or SaaS - Can design modular systems with evaluation logic and feedback loops
## How to apply Submit a short Loom video including: - Your background - An AI system you’ve built for multi-frame or sequential social content - Your experience with LLMs - Your answer to: Why does most AI-generated copy feel generic, and how would you design a system to fix it?
Applications without a Loom video or a relevant system example will not be considered.