Prompt Engineer
Overview
We are looking for a Prompt Engineer who specializes in designing, testing, and optimizing advanced prompts for AI systems. This role is not managerial or product-focusedit is a hands-on, iterative, technical writing role dedicated to building high-quality prompts that drive accurate and reliable results from models like OpenAI, Claude, and Gemini 2.5.
This is a full-time position, with most work taking place during PT hours.
Core Focus Areas
- Prompt Creation: Write, test, and refine complex prompts for structured data manipulation, analysis, and system interaction.
- Applied AI Use Cases: Develop prompts for tasks such as sentiment analysis on images and video, multi-step reasoning, and structured output formatting.
- Optimization: Continuously iterate on prompt performance to improve accuracy, consistency, and robustness across multiple models, while maintaining an emphasis on cost reduction.
- Validation & Auditing: Design and run processes to evaluate AI responses, including automated checks for output quality and correctness.
- Documentation: Maintain clear, reproducible records of prompt designs, experiments, and performance outcomes.
Key Responsibilities
- Craft and maintain production-grade prompts for internal and customer-facing systems.
- Explore and adapt emerging prompt engineering techniques for new model families.
- Run structured experiments to evaluate output quality, edge cases, and failure modes.
- Collaborate with engineers and analysts to align prompts with technical requirements.
Required Skills & Experience
- Proven prompt engineering experience, with portfolio/examples of complex prompts that manipulate data or control model outputs.
- Experience with prompting through API calls (not just through the web interfaces)
- Strong understanding of OpenAI, Claude, and Gemini 2.5 model behaviors, costs, context window lengths, tradeoffs, etc.
- Ability to analyze and structure information in a way that maximizes model reliability.
- Familiarity with evaluation techniques for LLM performance (e.g., output scoring, regression testing).
- Comfort working with technical systems and some exposure to programming (typescript preferred).
- Excellent written communication for precise prompt authoring and documentation.
- Excellent English verbal communication
What We Value
- Clarity: Writing that is unambiguous, structured, and logically sequenced.
- Creativity: Inventive approaches to achieving reliable outputs from ambiguous model behavior.
- Rigor: Systematic testing, validation, and refinement of prompts.
- Growth Mindset: Openness to new tools, techniques, and continuous improvement.
- Collaboration: Ability to work closely with engineers, analysts, and QA without taking on management responsibilities.
- Communications: Proactive in communications to ensure alignment with objectives.
Ideal Candidate
- Has a strong background in designing and refining prompts for real-world use cases.
- Learns quickly and thrives on experimenting with emerging AI capabilities.
- Is highly organized, with attention to detail in both writing and testing prompts.
- Has academic or professional background in: software engineering, linguistics, or marketing.
Additional Information
- Education: College degree preferred but not required. Demonstrated ability in AI prompt engineering is most important.
- Employment: Full-time preferred; part-time possible initially. Compensation based on experience.
How to Apply: Submit examples of prompts youve designed (or descriptions of systems youve worked on), along with your resume and a note on your experience working with OpenAI, Claude, or Gemini 2.5. to [Confidential Information]
