Agentic AI Engineer, AVP

Summary We are seeking an experienced Agentic AI Engineer to design, build, and evolve enterprise-grade Agentic AI platforms and solutions supporting investment management, investment research, investment performance, and related daily operations. The successful candidate will combine strong hands-on engineering capabilities with architectural judgment. The role will define and implement reusable Agentic AI capabilities across prompt engineering, memory, knowledge management, retrieval-augmented generation, agent orchestration, tool integration, evaluation, observability, security, and governance. The individual will work closely with platform engineering, data engineering, application teams, enterprise architecture, information security, model risk, and business stakeholders to transform business requirements into scalable, secure, explainable, and production-ready AI solutions.

Build production-grade Agentic AI solutions that automate and augment investment performance, investment research, financial analytics, document processing, and other operational workflows. Develop multi-agent and workflow-based solutions capable of decomposing complex business questions, retrieving relevant information, invoking approved tools and APIs, synthesizing results, and producing traceable outputs. Design and implement Retrieval-Augmented Generation solutions using structured and unstructured enterprise data, vector search, metadata filtering, document parsing, semantic retrieval, reranking, and source attribution. Integrate large language models and agent frameworks with enterprise applications, data platforms, databases, APIs, model gateways, and approved cloud services. Build reusable Agentic AI components and patterns, such as research agents, financial analytics agents, document-processing agents, evaluation agents, workflow agents, and governed tool-execution services. Establish engineering standards and guardrails for prompt versioning, context management, memory, agent state, tool permissions, error handling, fallback behavior, and deterministic workflow controls. Implement comprehensive evaluation frameworks covering answer quality, groundedness, retrieval relevance, tool-selection accuracy, task completion, latency, cost, safety, and regression testing. Deliver explainable and auditable solutions by preserving source references, generated queries, tool-call traces, execution history, model and prompt versions, and relevant decision records. Design solutions with appropriate identity, authentication, authorization, data access, encryption, logging, monitoring, and audit controls. Partner with Platform Engineering to align infrastructure, deployment, observability, CI/CD, secrets management, networking, resiliency, and production-support capabilities with solution requirements. Collaborate with data engineering teams on governed data ingestion, metadata management, data quality, schema evolution, lifecycle management, and Lakehouse integration. Review technical designs and code to ensure alignment with architectural principles, engineering standards, performance expectations, security requirements, and responsible AI practices. Work with enterprise architecture, security, compliance, privacy, model risk, and AI governance teams to support required reviews and production approvals. Diagnose production issues involving model behavior, retrieval, prompts, workflows, tools, application code, open-source libraries, and platform integrations. Create and maintain architecture diagrams, technical specifications, design guidelines, reusable implementation templates, operational runbooks, and architectural decision records. Evaluate emerging models, agent protocols, frameworks, and development tools, and recommend their controlled adoption based on measurable business and engineering value. Mentor engineers and contribute to