Overview

Core Responsibilities

Responsibilities

• Define reusable architectures and patterns for stateful agents, tool calling, memory, human-in-the-loop workflows, model orchestration, RAG, and vector retrieval.
• Design, build, deploy, and operate production-grade LLM applications, agent services, integrations, retrieval layers, and tool adapters using Python and modern engineering practices.
• Lead architecture and design reviews; make pragmatic trade-offs across quality, latency, reliability, cost, maintainability, data access, and security.
• Own the transition from prototype to production, including CI/CD, observability, incident response, rollback, and continuous improvement.
• Define evaluation and regression strategies for task success, factuality, groundedness, safety, robustness, and agent behavior.
• Establish tracing, monitoring, feedback, and service-level measures for quality, availability, latency, and cost per task.
• Design controls for privacy, PII, access management, prompt injection, data leakage, unsafe tool use, auditability, and human override.
• Partner with global AI, product, engineering, data platform, security, legal, and compliance teams to shape the roadmap and drive adoption of common AI building blocks.
• Mentor senior engineers and data scientists through design reviews, pair engineering, reference implementations, documentation, and operational playbooks.
• Evaluate models, frameworks, platforms, and third-party providers; lead technical due diligence and production-readiness assessments.

Qualifications

• Typically, 8 + years in software engineering, machine learning engineering, AI engineering, or a related discipline; equivalent depth will be considered.
• At least 3-4 years of delivering production AI, Generative AI, machine learning, or LLM-powered systems.
• Demonstrated ownership of production LLM applications, agentic workflows, RAG systems, or AI-enabled services from design through operation.
• Strong Python skills and experience building maintainable, tested, production-quality services and APIs.
• Hands-on experience with LangGraph, LangChain, or comparable agentic frameworks, including state, tools, memory, and human-in-the-loop patterns.
• Strong understanding of prompt/context engineering, embeddings, vector search, retrieval, and LLM evaluation.
• Experience with Git, Docker, CI/CD, cloud services, observability, and production deployment workflows.
• Working knowledge of PostgreSQL, Oracle, or comparable relational databases, including data quality, access, lineage, and performance considerations.
• Ability to lead technical decisions across teams without formal authority, communicate clearly in English, and operate with sound judgment in a controls-focused environment.

Nice to have

• Experience with AWS, GCP, Azure, Snowflake, Databricks, vector databases, model gateways, tracing, evaluation, or guardrail tooling.
• Experience operating AI systems in financial services or another regulated industry; knowledge of model risk, privacy, and secure software development.
• Background in MLOps, platform engineering, event-driven systems, distributed services, or high-availability production environments.
• Data science or machine-learning experience with PyTorch, scikit-learn, or equivalent frameworks.
• Practical experience with agentic coding tools such as Claude Code, Cursor, or comparable tools, with continued ownership of code quality, security, testing, and design.
• Shape a strategic agentic-AI capability in a global wealth-technology company.
• Own technical decisions from architecture through production operation across multiple products.
• Work across AI engineering, data science, product engineering, security, and financial-services technology.
• Join a diverse, international, and cross-functional engineering environment.