AI Research Scientist (PhD) | JPMorganChase AI Research

As a Research Scientist in AI Research, you will develop novel techniques, tools, and frameworks to model and solve complex, large-scale problems in the bank. You will build a close understanding of the challenges in practical AI applications and use it to shape new research directions that can shift the frontier. Your work will span early-stage innovation, rigorous evaluation, and production-scale delivery, in close collaboration with engineering and product teams.

Key Responsibilities

• Work on commercially oriented research projects with internal data scientists, applied engineering teams, and stakeholders across businesses such as Commercial & Investment Banking (including Markets), Asset & Wealth Management, and Consumer & Community Banking
• Formulate problems, generate hypotheses, develop new algorithms and models, run experiments and rigorous evaluations, communicate results, and deliver well-tested, high-quality code
• Contribute to high-impact business applications, reusable assets and products, and research initiatives
• Provide thought leadership in internal and external forums through white papers, publications, and presentations

Requirements

• Currently pursuing a PhD in Computer Science, Engineering, or a related field, with relevant AI/ML research experience
• Research publications in top-tier AI/ML venues, such as NeurIPS, ICML, ICLR, or highly regarded specialized venues like ICAPS and CRYPTO
• Ability to deliver modular, optimized, high-quality Python code with tests
• Deep understanding of fundamental AI/ML techniques and a strong grasp of the current state of the art in your area of expertise
• Practical experience with statistical data analysis, experimental design, and building meaningful benchmarks and metrics
• Effective verbal and written communication for technical and business audiences
• Motivation to make an impact on problems in financial services
• A strong research track record in at least one of these areas:
  • Synthetic data generation across modalities (tabular, images, time-series, etc.), synthetic personas, and agent simulations
  • Hybrid agent architectures combining LLMs with AI planning, formal logic, solvers, and neuro-symbolic approaches
  • Multimodal information discovery, deep research, and analysis across websites, news, documents, structured data, images, and audio
  • Foundation model training for structured data modalities such as tabular, graph, and time-series
  • Reinforcement learning, continual learning for agents, model post-training, and test-time adaptation
  • Advanced multi-agent systems, including coordination, negotiation, and mechanism design
  • Cryptography and secure multi-party computation, with applications to privacy-preserving ML and security in agentic systems

Skills

Technical Skills

• Python
• AI/ML research and algorithm development
• Statistical analysis, experimental design, and benchmarking
• Research specialization in one of the areas above

Soft Skills

• Curiosity, creativity, and resourcefulness
• Collaboration
• Communicating with technical and business audiences
• Thought leadership through publications and presentations

Nice to Have

• Experience with ML libraries (e.g., PyTorch, TensorFlow/Keras, HuggingFace Transformers) and agent frameworks (e.g., LangGraph, Google ADK), depending on specialization
• Familiarity with formal verification tools and languages (e.g., Coq, Isabelle, Lean, TLA+, Alloy, Z3, PDDL)
• Practical software development experience in collaborative settings, such as open-source projects or industry
• Proven ability to turn business requirements into technical problem formulations and deliver novel solutions into production

Benefits

• Competitive total rewards, with possible commission-based pay and/or discretionary incentive compensation for eligible roles
• Comprehensive health care coverage and on-site health and wellness centers
• Retirement savings plan
• Backup childcare
• Tuition reimbursement
• Mental health support
• Financial coaching