Prompt Systems Engineer

Zipporin is seeking an experienced Prompt Systems Engineer to join our AI Engineering team in New York, NY on a hybrid basis. In this role, you will bridge the gap between foundation models and production software by designing deterministic, scalable, and resilient prompt architectures and agentic workflows.Key ResponsibilitiesArchitect, implement, and maintain advanced prompt pipelines, dynamic context-retrieval systems, and multi-agent workflows across multiple frontier LLMs.Develop rigorous, automated evaluation frameworks (evals) to continuously monitor model accuracy, hallucination rates, bias, and output schema conformity.Collaborate with backend and ML engineers to integrate prompt chains with vector stores, relational databases, and low-latency production APIs.Implement semantic caching, context compression, and token optimization strategies to minimize inference latency and operational costs.Stay abreast of generative AI breakthroughs, systematically benchmarking new foundation models, embedding techniques, and prompt optimization paradigms.Qualifications & RequirementsBachelor's or Master's degree in Computer Science, Computational Linguistics, AI, or equivalent practical experience.3+ years of professional software engineering experience, with strong proficiency in Python and asynchronous API architectures.Proven track record building and deploying production-grade LLM applications using frameworks like LangChain, LlamaIndex, DSPy, or Semantic Kernel.Hands-on experience with Retrieval-Augmented Generation (RAG) architectures and modern vector databases such as Pinecone, Weaviate, or Qdrant.Strong understanding of foundation model mechanics, context window optimization, few-shot prompting, and automated testing frameworks.