We are the Armenian Artificial Intelligence Company (AAIC): a collective of scientists, engineers, designers, and writers. We work globally, creating unique software for diverse clients—from devices that assist professional chefs to city simulations and narrative games.

As we launch new narrative projects, we're expanding our writers' room and seeking a prompt engineer with a passion for fiction to support our writing team.

What we're looking for:

• Hands-on experience with LLM text generation: you regularly work with large language models like ChatGPT or DeepSeek and know how to craft prompts that produce stylistically and factually consistent texts.
• Hands-on experience using LLM for non-generative tasks: you apply LLM for classification or input validation purposes and know how to craft prompts that produce reliable machine-readable output.
• Hands-on experience with coding: you can write simple Python or TypeScript scripts to automate your tasks when needed.
• Hands-on experience in writing: you've created essays, novels, stories, academic papers, or articles—we accept any form of written work in any language (except social media posts, technical documentation, or project briefs).
• You are fluent in English and have excellent written communication skills. Fluent Russian is a big plus.

This position is fully remote with an expected workload of 30–40 hours per week at a rate of $10–15 per hour.

What we will NOT be looking at:

• Years of experience.
• Big names in your CV.
• We'll read the CV if you send it, but it doesn't matter that much in general.

To apply, email Niko at [email protected] with the following:

• An LLM prompt focused on text generation (e.g., generating documents, chat messages, code)
• An LLM prompt focused on classification or validation (e.g., sentiment analysis)
• A piece of practical code that you wrote yourself (doesn’t need to be industrial-grade)
• A sample of your long-form fiction writing (can be in any language).
• A description of the most interesting project you've worked on (doesn't need to be commercial or large-scale).