Description

You will engineer, test, and deploy advanced prompts for large language models.

Responsibilities

• Design, refine, and test prompts using few-shot/zero-shot, chain-of-thought, and schema-guided output techniques.
• Build and maintain modular prompt libraries, engineering context using RAG, memory, and tool-calling for complex tasks.
• Evaluate prompt performance via A/B testing and metrics (BLEU/ROUGE, human eval), analyzing failure modes like hallucinations.
• Collaborate with teams to align prompts with business needs, integrating them into applications via APIs and workflows using LangChain/LlamaIndex.
• Automate prompt experiments, data preprocessing, and evaluation harnesses in Python, while monitoring production LLM drift.

Required Skills

• 2+ years hands-on prompt engineering with production LLMs.
• Proficiency in Python for API integration, scripting, and evaluation.
• Experience with LLM platforms including GPT, Claude, and Vertex AI.
• Familiarity with LLM frameworks such as LangChain and Hugging Face.
• Strong grasp of tokenization, context windows, temperature, and model behavior.
• 2+ years of software development experience building production applications.
• Ability to decompose complex tasks and design rigorous evaluation frameworks.

Key Skills

Education

Bachelor's degree