Jobs → Engineering → Gen AI Engineer

About the Role We are looking for a Gen AI Engineer to join our team in NY. This role focuses on designing and evolving reusable GenAI workflow primitives and services used across Institutional Securities workflows. Key Responsibilities • Design and evolve reusable GenAI workflow primitives and services used across Institutional Securities workflows • Develop AI-powered assistants embedded into core Institutional Securities applications, leveraging agentic and tool-driven workflows • Define and guide GenAI architecture decisions, including model selection, orchestration patterns, and evaluation strategies • Establish and evolve LLMOps practices, including evaluation harnesses, prompt/version management, monitoring, and regression testing • Design and implement controls for entitlements, data security, and PII handling, including usage of open-source models in regulated environments • Partner with business and platform teams to drive adoption of shared GenAI capabilities across systems and workflows Requirements • 2+ years of hands-on experience building and operating GenAI systems in production • 7+ years of full-stack or platform engineering experience, with strong proficiency in Python • Proven experience designing and operating LLM-based systems using patterns such as RAG, tool/function calling, agentic workflows, and structured outputs • Strong expertise in LLMOps, including evaluation frameworks, prompt/version management, regression testing, observability, and production reliability • Experience building AI-first document ingestion and extraction pipelines with measurable quality and accuracy • Experience with coding agents (Claude code, Codex, AMP, CoPilot) • Advanced experience in retrieval systems, including multi-stage pipelines, vector search, re-ranking, metadata filtering, and evaluation metrics (e.g., recall/precision tradeoffs, MRR, NDCG) • Practical experience debugging and stabilizing systems through real-world failure scenarios, including model regressions, prompt drift, retrieval degradation, and data quality issues