Only W2 ABOUT THE ROLE Own prompt design and evaluation for LLM-powered chatbots, agents, and RAG systems. You will turn business requirements into reliable prompt and agent workflows and prove they work with measurable results. RESPONSIBILITIES Craft, test, and refine prompts that maximize accuracy, relevance, and consistency Build evaluation frameworks: A/B testing, benchmarking, golden datasets, and rubric-based scoring Apply deep knowledge of LLM behavior: context windows, tokenization, and fine-tuning versus prompting trade-offs Partner with product, engineering, and data science to design prompt and agent workflows Implement prompt security and safety: jailbreak prevention, bias mitigation, and guardrails Maintain prompt versioning and regression testing REQUIRED QUALIFICATIONS: 5+ years in software, data, or applied AI roles, with minimum 1+ year of hands-on LLM prompt work in production in 5+ yrs of experience Strong Python scripting and API integration Hands-on experience with GPT and Claude, and orchestration tools such as LangChain Demonstrated experience designing evaluation methods and iterating on results Working knowledge of RAG and agent patterns, including tool calling Data analysis skills to interpret evaluation results Apply To This Job