Applied Machine Learning Scientist - Vice President

JPMorganChase • Palo Alto, CA

1d

About The Position

As an Applied Machine Learning Scientist, you will lead the development of scalable, production-grade machine learning solutions across natural language processing, speech recognition, recommendation systems, information retrieval, and agentic AI. You will design and productionize generative AI capabilities, including LLM-powered RAG, tool/function-calling agents, and structured generation, to automate complex workflows and improve customer experiences. You will collaborate with product, engineering, and control partners to turn ambiguous problems into measurable goals, deliver robust models, and operate them reliably in production. The role requires deep learning and transformer modeling expertise, hands-on fine-tuning and evaluation experience, analytical thinking, motivation to learn and experiment with new innovations, and participation in a knowledge-sharing culture. The position is subject to Section 19 of the Federal Deposit Insurance Act. An employment offer is contingent on JPMorganChase’s review of criminal conviction history, including pretrial diversions or program entries.

Requirements

• Master’s degree with 7+ years, or Ph.D. with 4+ years, of hands-on industry experience building and deploying production machine learning systems in NLP, information retrieval, recommendation systems, and/or generative AI.
• Good understanding of current NLP advances, including transformer architecture, knowledge distillation, transfer learning, and representation learning.
• Applied generative AI experience with LLMs and ability to fine-tune and deploy SLMs for targeted use cases; familiarity with prompt design, grounded generation, and RAG.
• Experience scaling LLM systems, including caching, batching, prompt/version governance, and evaluation harnesses.
• Strong foundation in machine learning, deep learning, and statistical modeling, including model evaluation and error analysis.
• Solid understanding of information retrieval concepts such as indexing, ranking, dense/sparse retrieval, and re-ranking, and/or recommendation systems.
• Ability to design experiments, establish strong baselines, choose meaningful metrics, and rigorously evaluate model performance.
• Scientific thinking and ability to invent and work independently as well as in highly collaborative teams.
• Proficiency in Python and common ML libraries such as PyTorch or TensorFlow, Hugging Face, and scikit-learn; ability to write production-quality code.
• Ability to collaborate across product, engineering, and control functions.
• Strong written and spoken communication skills.
• Strong analytical thinking, motivation, passion for machine learning, and desire to learn.

Nice To Haves

• Five years of hands-on experience with virtual assistant model development and optimization.
• Experience orchestrating multi-agent teams with supervisor agents, debate or consensus mechanisms, and role-specialized toolkits for complex enterprise tasks.
• Experience building agent governance and evaluation suites, including red-teaming, adversarial tests, safety scorecards, and prompt/tool regression suites.
• Experience with reinforcement learning or bandits, preference optimization, or human feedback loops for personalization.
• Experience in regulated finance domains and working with risk/control processes.
• Experience with MLOps/LLMOps, including model CI/CD, monitoring and alerting, model versioning, pipeline evaluation, and rollback strategies.
• Experience with A/B experimentation and data- and metric-driven product development.

Responsibilities

• Lead and deploy advanced machine learning systems across NLP, speech recognition, recommendation systems, and information retrieval.
• Design and build agentic AI systems for multi-step workflows, including tool/function calling, multi-agent orchestration, planning, grounding, and safety guardrails.
• Apply reinforcement learning approaches, including policy optimization, bandits, and RLHF-style methods where appropriate, to improve personalization, dialog policies, and sequential decision-making systems.
• Fine-tune and adapt LLMs and SLMs using PEFT methods such as LoRA, AdaLoRA, and IA3, as well as distillation and quantization; optimize quality, latency, cost, and production readiness.
• Select and innovate on machine learning strategies for banking problems.
• Analyze and evaluate the ongoing performance of developed machine learning systems.
• Collaborate with Business, Technology, Product Management, Design, Analytics, and Model Governance partners to deploy solutions into production.
• Build domain understanding, identify high-impact opportunities, ensure responsible AI use, and drive measurable customer experience, automation, accuracy, and efficiency outcomes.
• Implement privacy, safety, and security controls for generative AI systems, including PCI handling and redaction, policy checks, jailbreak resistance, and auditability.
• Invest independent time in learning, researching, and experimenting with new innovations, and contribute to knowledge sharing.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior