Job Description
We are looking for a highly creative and analytical Prompt Engineer to join our innovative team in Jersey City, NJ. In this role, you will define the interaction between human users and our advanced AI systems, crafting the precise language that drives our Large Language Models (LLMs). We provide a premium remote setup with top-tier equipment, including high-performance GPUs and workstations, ensuring you have the tools needed to push the boundaries of AI capabilities.
Why You’ll Love This Role:
- Equipment Provided: We provide a state-of-the-art MacBook Pro M3 Max, high-speed internet, and access to cloud GPU clusters for testing.
- Flexible Work: Hybrid model with a collaborative office in Jersey City and remote flexibility.
- Impact: Shape the future of AI interaction directly through your language expertise.
Responsibilities
- Design, iterate, and optimize complex prompts for various LLMs (e.g., GPT-4, Claude, Llama 3) to achieve specific outcomes.
- Develop and maintain a scalable prompt library to ensure consistency and reliability across the organization.
- Collaborate closely with Data Scientists and Product Managers to fine-tune models based on user feedback and performance metrics.
- Analyze AI outputs to identify bottlenecks and implement strategies to improve accuracy and reduce hallucinations.
- Document prompt engineering methodologies and create training materials for internal teams.
- Stay ahead of industry trends in AI, NLP, and machine learning to inform strategic decision-making.
Qualifications
- 3+ years of experience in technical writing, copywriting, or prompt engineering.
- Strong command of the English language with exceptional attention to nuance, tone, and clarity.
- Deep understanding of transformer models, tokenization, and the underlying mechanics of NLP.
- Experience with Python scripting for automating prompt testing and logging results.
- Ability to translate complex user requirements into precise, executable prompts.
- Portfolio or case studies demonstrating successful prompt optimization or LLM integration.