Machine Learning Engineer

We are looking for a driven and innovative  Machine Learning Engineer  to join our Global Retirement & Wealth AI team! Within our AI and Generative AI initiatives, you will compose, build, and scale production‑grade GenAI solutions. These solutions improve participant outcomes, advisor productivity, and operational efficiency across our Retirement and Wealth businesses. Your work will power experiences such as participant and advisor copilots, RAG over plan and product documents, personalization and retirement readiness guidance, workflow automation and intelligent compliance enablement!

Position Responsibilities

Build GenAI Products End-to-End:
Design, implement, and product ionize LLM‑powered applications (APIs, microservices, and UI‑backed services) including RAG pipelines, tool/Function‑calling, agentic workflows, and prompt orchestration for participant, plan sponsor, advisor, and operations use cases.

Data & Retrieval Engineering:
Partner with data engineers and SMEs to source, model, and optimize structured and unstructured data (plan documents, product guides, call notes, knowledge bases). Implement embedding pipelines, chunking strategies, retrieval optimization, and vector search (e.g., Azure AI Search, Redis/Mongo

DB/Lance

DB).

Modeling & Optimization:
Apply classical ML and GenAI techniques (prompt engineering, fine‑tuning, RAG, reranking, guardrails) to improve accuracy, latency, cost, and hallucination control.

MLOps & LLMOps:
Ship reliable services with CI/CD, infrastructure‑as‑code, model/prompt versioning, MLflow/experiment tracking, observability, canary releases, and automated evaluation suites (offline & online A/B).

Security, Privacy, and Compliance:
Implement and document controls for PII/financial data, RBAC, prompt injection defenses, content filtering, red‑teaming, and model risk governance aligned to regulatory expectations (e.g., auditability, explainability, record‑keeping).

Partner

Collaboration:

Translate business goals into technical plans with Product, Operations, Contact Center, Distribution, Compliance, Risk, and Legal. Convert requirements into robust builds and SLAs; standardize methods into engineering guidelines reusable across teams.

Continuous

Innovation: Know the latest on LLM and retrieval research, model choices, evaluation techniques, and platform capabilities and pragmatically apply them to Retirement & Wealth use cases at scale.

Required Qualifications

Education:

Bachelor’s or Master’s in Computer Science, Data Science, Engineering, Mathematics, or a related quantitative field.

Experience:

4+ years building ML/AI solutions, including 2+ years hands‑on with Generative AI (RAG, prompt engineering, function/tool calling, agentic patterns). Consistent track record of shipping production systems with measurable business value.

Programming & Frameworks:
Strong Python and SQL; experience with Lang Chain, Lang Graph, Semantic Kernel, CrewAI, or ADK; familiarity with Hugging Face, PyTorch, and modern embedding/reranker stacks.

Cloud & Data:
Practical experience on Azure (preferred) and Databricks/Spark, Delta Lake/Unity Catalog, feature stores, API development, containerization (Docker, Kubernetes), and event/messaging (e.g., Kafka/Event Hubs).

RAG & Retrieval:
Hands‑on with embedding models, chunking, metadata enrichment, vector databases/search, hybrid search and retrieval evaluation/telemetry.

MLOps/LLMOps: CI/CD (e.g., Git Hub Actions/Azure Dev Ops), model and lifecycle management, experiment tracking (MLflow), observability, evaluation harnesses, and optimization of expenses and response times.

Communication:
Excellent ability to translate complex ML/LLM concepts into business outcomes for both technical and non‑technical partners; clear documentation and design articulation.

Preferred Qualifications

Safety & Governance:
Exposure to model risk management, prompt and content safety guardrails, adversarial testing/red‑team, and responsible AI practices.

Fine‑Tuning:
Practical experience with LoRA/PEFT, supervised fine‑tuning, or instruction‑tuning pipelines; evaluation with task‑specific metrics and human‑in‑the‑loop review.

Platform & Tooling:
Familiarity with Azure OpenAI,…