Join a world-class Applied AI/ML organization at JPMorgan Chase and help shape how teams across the firm use data science, machine learning, and Generative AI to solve real business problems. In this shared services role, you’ll support Consumer & Community Banking (CCB) Control Management Shared Services by delivering horizontal capabilities that strengthen how Control Managers operate day-to-day across Consumer & Community Banking businesses and functions (e.g., Auto, Home Lending, Credit Card, Consumer Banking, Business Banking, Operations, Branch Review, and ICB), spanning core activities like ongoing risk monitoring, process and regulatory understanding, metric/breach review, and building a holistic view of risks, controls, issues, action plans, applications, and intelligent automation in the control environment.
As a Senior Associate in Applied AI/ML (Shared Services), you will design and deploy predictive ML, advanced analytics, and GenAI/LLM agentic solutions—systems that orchestrate tools, workflows, and large language models within business processes—to create reusable services that scale across the Control Management lifecycle: maintaining risk assessment structures and tagging, supporting legal/regulatory change and obligation mapping, improving risk assessment and MRI alignment, enabling control design/testing and sustainable monitoring, accelerating issue identification/root-cause/action-plan tracking and validation, and strengthening governance, committees, scorecards, and reporting.
Job Responsibilities
- Design, develop, and deploy predictive ML, advanced analytics, GenAI/LLM, and agentic AI solutions for complex business problems in shared services.
- Build and integrate agentic workflows (tool use, RAG, routing/planning, structured outputs, evals/guardrails) into end-to-end business processes to deliver context-aware insights and automation.
- Prototype AI-enabled approaches quickly, then harden successful prototypes into reusable, production-ready services with measurable outcomes.
- Own end-to-end model delivery: dataset manipulation/feature engineering, training, validation, evaluation, deployment, and iteration.
- Design, deploy, and operate production ML pipelines and services (batch/real-time), including logging/metrics, monitoring, retraining/refresh strategies, and reliability/cost/latency improvements.
- Partner with product, engineering, and risk/controls stakeholders to define requirements, align on success metrics, and drive adoption.
- Apply responsible AI, governance, and compliance-aligned practices throughout the model and agent lifecycle; share best practices and contribute reusable templates/libraries.
Required qualifications, capabilities, and skills
- Bachelor’s degree in data science, computer science, statistics, mathematics, or a related technical field (or equivalent practical experience).
- 5+ years experience or demonstrated ability to set up and deploy AI/ML solutions end-to-end (prototype → production or production-like), shown through prior roles, internships, research, or substantial projects.
- Strong Python proficiency for data analysis, modeling, and production-grade implementation; solid dataset manipulation and feature engineering skills.
- Hands-on experience building, evaluating, and deploying predictive models and analytics solutions (e.g., classification/regression, NLP) using common ML/deep learning libraries (e.g., PyTorch, TensorFlow, scikit-learn).
- Required agentic AI experience: built and deployed LLM-enabled agentic workflow (e.g., RAG + tool/function calling, routing/planning, structured outputs) with an evaluation approach (test set, regression tests, human review, or similar).
- Experience designing, deploying, and operating production ML/LLM pipelines or services, including basic MLOps practices (versioning, CI/CD for ML, monitoring/alerting, incident hygiene).
- Working knowledge of modern deployment environments: cloud (AWS/Azure/GCP) and/or containerized/distributed compute (e.g., Kubernetes).
- Strong communication and stakeholder partnership skills; ability to translate business problems into measurable technical outcomes and explain results to diverse audiences.
Preferred qualifications, capabilities, and skills
- Advanced education & thought leadership: Master’s or PhD in a quantitative field; publications, patents, or meaningful open-source contributions in ML/GenAI.
- Advanced agentic/GenAI maturity: scaled agentic systems beyond a single use case; strong LLM evaluation discipline (golden sets, automated regression, quality dashboards) and guardrail patterns.
- Scale/performance & data ecosystems: GPU/inference optimization (e.g., Triton, profiling), big data processing and cloud data services; exposure to RL or other advanced ML methods.
- Specialized ML domains & regulated environments: search/ranking, recommenders, graph ML/knowledge graphs; experience in financial services or other regulated industries and comfort operating within governance expectations—especially for regulatory/change management workflows.
What You’ll Build in Shared Services
- Reusable agent frameworks and patterns (routing, tool-use, workflow orchestration, safety controls) that multiple teams can adopt.
- LLM-powered capabilities embedded in business processes (summarization, classification, decision support, workflow automation) with measurable quality and risk controls.
- Deployed models supporting regulatory and change management (e.g., obligation/change classification and tagging, QA/routing, impact triage, and audit-ready decision support) integrated into workflows with monitoring and governance.
- Evaluation and monitoring foundations (golden sets, automated regression tests, drift/quality dashboards) that standardize how AI is operated at scale.