Director, AI Platform & Agent SDK

J.P. Morgan; New York (NY); On-site; USD 210,000 - 320,000; Full time

Benefits offered by this job

Health care coverage

On-site health and wellness centers

Retirement savings plan

Tuition reimbursement

Mental health support

Financial coaching

Job summary

JPMorganChase in New York seeks an Applied AI/ML Director to shape an Agent Builder Platform. You will own core SDKs, build scalable agent capabilities, and drive adoption across product and engineering teams.

This hands-on leadership role requires delivering robust, production-ready components end-to-end. You will collaborate with data, platform, and application teams to ensure solutions are secure, scalable, and measurable against business metrics.

Qualifications

• 10+ years applied experience in AI/ML
• Experience building large-scale ML systems and platform services
• Strong software engineering from design to operation
• Proficiency with PyTorch or TensorFlow
• Agentic/generative AI system design and evaluation

Responsibilities

• Architect and implement core Agent SDK capabilities and reference implementations
• Build specialized agents and reusable components improving reliability and observability
• Translate agentic/AI techniques into scalable platform features
• Design evaluation approaches for agent behavior including latency, cost
• Develop model-serving/workflow patterns for agentic systems
• Collaborate with product/engineering to align platforms with metrics
• Drive technical decisions with clear design and trade-offs
• Improve platform performance via profiling and optimization

Job description

JPMorganChase in New York seeks an Applied AI/ML Director to shape an Agent Builder Platform. You will own core SDKs, build scalable agent capabilities, and drive adoption across product and engineering teams.

This hands-on leadership role requires delivering robust, production-ready components end-to-end. You will collaborate with data, platform, and application teams to ensure solutions are secure, scalable, and measurable against business metrics.