JPMorgan Chase is looking to solve real-world challenges impacting the company, customers, and communities by anticipating new and emerging risks and growing its business responsibly. The Data Scientist Vice President will contribute to this by devising and developing Proofs of Concept (POCs) and deployable models using AI/Gen AI/ML techniques.
Requirements
- 6+ years of related experience in Python, R or Scala
- Demonstrable theoretical and application knowledge of AL/ ML, Gen AI and Statistical Models
- Demonstrable hands-on experience with Transformer or other deep learning architectures in real applications
- Demonstrable hands-on experience with using or fine tuning multimodal LLM in real business applications with scale and performance
- Demonstrable hands-on experience and familiarity with any or all of the following packages, algorithms, and/or alternatives, including Graph Learning Packages : (NetworkX, Torch-Geometric, Graphframes, Graphistry),ML Packages (Pandas, Scikit-Learn, XGBoost, catboost, lightgbm, automl, Optuna, Hyperopt), Visualization Packages (Matplotlib, Seaborn, Geopandas), Algorithm (Ensemble Louvian / Hierarchical Clustering, Label Propagation, Connected Component Analysis, Graph Neural net (Graph Attention Network), Page Rank, Centrality Analysis, Tree based Analysis, Outlier Detection Methods, Zero Shot/ Few Shot learning)
- Demonstrable experience with graph analytics, graph-based learning, and graph representation/visualization
- Experience in graph Database: TigerGraph, Neo4j
- Experience in Query Language: Hive, Cypher (Graph Query Language)
- Experience in developing and operationalization of data pipelines
- Familiarity and experience of assimilating large amounts of data from multiple databases and utilize them for creating actionable outcome
- Hands-on professional experience in software development especially with analytical & computationally intensive systems, digital transformations leveraging cloud technologies (AWS, GCP, Azure, Databricks etc.).
- Experience with EKS or AKS provisioning is a plus
- Working knowledge of C/C-Sharp/C++ or others is a plus
Responsibilities
- Devising and developing Proofs of Concept (POCs) and deployable models using AI/Gen AI/ ML techniques, algorithms and other statistical and numerical methods.
- Extract and work with large volumes of data (both structured and unstructured) from multiple sources, transforming it into an analysis-ready format to develop the data pipeline.
- Independently formulate methodologies, and quantitative and analytical tasks, from business problems.
- Analyze complex/unstructured data to understand the business problem and use case
- Analyze business requirements, design, and develop appropriate methodology
- Develop deployable, scalable and effective models/ analytical methods as part of technology managed system or as a self-served application of a business user
- Prepare technical documentation of quantitative models for internal model risk and governance review
Other
- Work collaboratively and creatively with other data scientists, technology partners, risk professionals, model validation teams, etc.
- Adhering to a standardized analysis and project methodology; and documenting quantitative analysis
- Experience with processes, controls and governance of a highly regulated environment
- Self-starter and strong influencing skills with strong communication skills
- Real life exposure to Agile SDLC, ModelOps and /Or Design Thinking is desirable.
- Experience in financial services industry especially in Operational Risk Management, Anti-Money Laundering & Know Your Customer, Trade Surveillance model development
