At the heart of JP Morgan’s AI transformation is the Chief Analytics Office (CAO) — the team responsible for driving the firmwide adoption of artificial intelligence and advanced analytics. The CAO leads the strategy, governance, and delivery of AI/ML products across the company, ensuring that innovation is balanced with responsibility, security, and ethical best practices. By overseeing the build, adoption, and maintenance of AI solutions, the CAO empowers every line of business to leverage AI/ML at scale and unlock new value for clients and stakeholders.
As a Generative AI Executive Director within our CAO organization, you’ll be at the forefront of building and optimizing production-grade LLM systems (LLM Suite) that serve hundreds of thousands of professionals every day. You’ll architect and scale robust, reusable APIs and agentic workflows that process millions of documents and automate complex financial tasks. Collaboration is key: you’ll work closely with teams across ML Engineering, Product Management, and Cloud Engineering to deliver solutions that are measured, budgeted, and built for real business impact.
Your technical decisions will directly shape how JP Morgan leverages AI at enterprise scale. You’ll ensure LLM Suite is designed for reliability, scalability, and performance — enabling other teams to build on top of our infrastructure and accelerate innovation firmwide. This is not a research sandbox; it’s production infrastructure with executive visibility and measurable ROI.
If you’re ready to lead engineering teams, ship production AI, and make decisions that influence the future of financial technology, we invite you to join us and help define the next era of enterprise AI at JP Morgan.
Job Responsibilities
- Architect and deliver production LLM based systems (text, image, speech, video) powering mission-critical LLM Suite products.
- Own end-to-end delivery, performance, and continuous improvement of individual LLM Suite products.
- Bridge advanced AI research with robust engineering to build innovative, production-ready solutions.
- Drive results with an entrepreneurial mindset in a fast-paced, high-impact environment.
**Required qualifications, capabilities, and skills **
- PhD or equivalent experience in Computer Science, Mathematics, Statistics, or a related quantitative discipline.
- Extensive hands-on experience as an individual contributor in ML engineering, with a proven track record of shipping production AI systems.
- Deep expertise in NLP, Computer Vision, and/or Multimodal LLM algorithms, with a strong foundation in statistics, optimization, and ML theory.
- Practical experience implementing distributed, multi-threaded, and scalable applications using frameworks such as Ray, Horovod, DeepSpeed, etc.
- Exceptional communication skills, able to convey complex technical concepts and build trust with stakeholders at all levels.
**Preferred qualifications, capabilities, and skills **
- Advanced proficiency in designing and deploying production ML pipelines using DAG frameworks, including custom operator development and pipeline optimization.
- Expertise in architecting and implementing high-throughput, low-latency microservices with gRPC, REST, and GraphQL, including protocol buffer schema design, streaming endpoints, and load balancing.
- Hands-on experience with parameter-efficient fine-tuning (LoRA, QLoRA, IA3), model quantization (INT8, FP16, GPTQ), and quantization-aware training for LLMs at scale.
- Deep knowledge of distributed training strategies (data/model/pipe parallelism), memory optimization, and inference acceleration for large-scale multimodal models.
- Experience with advanced agentic workflow orchestration, including multi-agent coordination, stateful task management, and integration with enterprise event-driven architectures.
#LI-RB1