Principal Data Engineer

As Principal Data Engineer at Anaplan, you will shape the data backbone for AI-powered planning workflows. You’ll drive end-to-end GenAI features, from data ingestion and model integration to monitoring and deployment, enabling real-time AI capabilities for business users. You’ll design retrieval layers for RAG, build knowledge graphs, and collaborate with data scientists to productionise ML models. This role sits at the intersection of data engineering and AI, offering impact at scale within a winning culture that emphasizes customer success.

• Design and build the retrieval layer powering RAG and agentic workloads, including vector/graph databases, hybrid search, and knowledge graphs
• Develop end-to-end GenAI features including backend API services, model integration, monitoring, evaluations and deployments
• Engineer feature and context pipelines balancing batch and streaming patterns to feed forecasting and anomaly-detection models, collaborating with data scientists to productionise algorithms
• Build the data plane for evaluation with rigorous frameworks to monitor, measure, and improve GenAI feature quality, accuracy, latency, and user satisfaction
• Collaborate with data scientists to productionise ML models and forecasting algorithms

• Extensive background in Data Science Engineering with principal-level technical leadership
• Hands-on experience building and shipping AI/ML products in production
• Deep practical experience with LLM-based systems: RAG architectures, embedding pipelines, prompt and response logging, evaluation frameworks
• Hands-on expertise with vector databases, graph databases, and knowledge graphs
• End-to-end exposure in model lifecycle development, including training and deploying ML models in production
• Deep knowledge of LLM APIs, prompt engineering, and conversational AI patterns
• Proficiency in Python and modern software development practices (testing, code review, CI/CD)

• leadership
• collaboration
• problem-solving
• LLM-based systems (RAG, embeddings, prompt logging, evaluation frameworks)
• Vector databases (e.g., Pinecone, Weaviate, Qdrant)
• Graph databases and knowledge graphs