Zipporin is seeking a talented and disciplined Prompt Systems Engineer to join our core AI engineering organization based in Seattle, WA (hybrid). In this role, you will be instrumental in bridging generative AI research and enterprise-scale software engineering by architecting, optimizing, and systematizing prompt workflows, agentic chains, and retrieval-augmented generation (RAG) pipelines.
Key Responsibilities:
- Design, calibrate, and productionize robust prompt architectures and multi-step agentic pipelines across diverse LLMs.
- Construct automated benchmarking and regression testing suites to measure prompt output quality, hallucination rates, latency, and token economics.
- Integrate vector search, hybrid retrieval strategies, and context injection techniques into production backend services.
- Establish guardrails, content moderation rules, and defensive prompt patterns to secure against prompt injection and data exfiltration.
- Collaborate closely with product managers and backend developers to transition experimental AI prototypes into scalable, high-throughput microservices.
Requirements:
- 3+ years of hands-on software development experience, including at least 1-2 years focused on LLM integration, prompt tuning, and generative AI systems.
- Advanced proficiency in Python and hands-on experience with orchestration frameworks like LangChain, LlamaIndex, or Semantic Kernel.
- Demonstrated experience deploying production RAG pipelines utilizing vector databases such as Pinecone, Qdrant, or Weaviate.
- Familiarity with quantitative evaluation tooling and metrics (e.g., Ragas, DeepEval, BLEU/ROUGE, LLM-as-a-judge).
- Strong foundational knowledge of software design principles, version control, and CI/CD best practices.