AI Prompt Engineer

CDIT LLC is seeking a highly creative, technically proficient, and detail-oriented AI Prompt Engineer to join a growing AI team supporting a federal border security mission. This role is responsible for designing, testing, and optimizing prompts for Large Language Models (LLMs) to produce high-quality, reliable outputs across a range of mission use cases — translating complex requirements into effective AI interactions, within an Agile delivery environment.

•     Prompt Design & Optimization — Develop, refine, and iterate on prompts for generative AI models (LLMs) to elicit desired behaviors and outputs.

•     Iterative Testing & Analysis — Conduct structured testing to compare prompt variations and drive continuous improvement.

•     Collaboration — Work closely with cross-functional teams to integrate prompts that streamline manual processes.

•     Fine-Tuning & RAG Support — Provide input on model fine-tuning and Retrieval-Augmented Generation (RAG) strategies.

•     Problem Solving — Debug unexpected model behaviors, identifying root causes in prompt structure or data context.

•     Must be a U.S. Citizen with the ability to pass a federal background investigation (criminal, drug use, and misconduct checks).

•     Bachelor's degree in Computer Science, Software Engineering, IT, or related field (equivalent experience considered)

•     0–3 years of professional software development experience or relevant academic project work

•     1+ year of hands-on experience with LLMs (e.g., OpenAI GPT series, Llama, Gemini)

•     Proficiency in at least one programming language (Python, JavaScript/TypeScript)

•     Familiarity with prompt engineering techniques (few-shot, chain-of-thought, structured outputs, function calling)

•     Strong, data-driven problem-solving skills

•     Strong written and verbal communication skills

•     Exposure to RAG concepts, embeddings, and vector databases

•     Understanding of AI safety considerations (prompt injection, bias, data privacy)

•     Exposure to Agile/Scrum environments