Expert AI Engineer

iPipeline is a global market leader that provides innovative software solutions for the life insurance and financial services industries. The Expert AI Engineer will own complex technical work within GenAI and AI platform services, focusing on designing and optimizing retrieval pipelines and inference systems while guiding less-experienced team members.

Responsibilities:

• Implement and optimize RAG pipelines, embeddings workflows, and LLM integration patterns
• Contribute to scalable, low-latency inference architecture
• Develop and optimize components of real-time and batch inference pipelines that support document processing, portfolio insights, and decision-support use cases. Improve pipeline performance, reliability, and integration quality across the team’s GenAI workflows
• Design sections of ingestion, transformation, and indexing pipelines for vector stores and hybrid retrieval. Design data curation processes and retrieval corpora, working with domain experts to ensure content quality, provenance, and relevance for GenAI use cases
• Lead improvements in cost-efficiency, performance, and reliability across components
• Apply secure-by-design principles and contribute to drafting responsible AI guidelines
• Design and maintain prompt templates, orchestration flows, and model configurations, establishing patterns for versioning, rollback, and auditability across the team’s GenAI workflows
• Design and implement guardrail patterns (e.g., safety classifiers, content filters, policy checks) that consistently mitigate harmful or non-compliant outputs across multiple GenAI features
• Design small-scale evaluation modules measuring grounding, factuality, consistency, and quality
• Build automated test harnesses and evaluation scripts for model iterations
• Reduce hallucinations by applying grounding strategies and structured response patterns
• Partner with SMEs to translate qualitative evaluation findings into engineering revisions. Analyze complex evaluation data to inform design decisions
• Design evaluation tests and datasets focused on safety, bias, and compliance, and analyze results to propose concrete model or workflow changes
• Lead design and implementation of core components of GenAI services and pipelines
• Participate in architectural discussions and propose improvements within the product area
• Write modular, reusable code used broadly by team members
• Contribute in technical design reviews and code reviews. Mentor less experienced team members on coding standards, testing, and best practices
• Conduct detailed performance troubleshooting and optimize system bottlenecks