Director, Data Science

Fidelity Investments•Durham, NC

79d•Hybrid

About The Position

Builds algorithms using programming languages (Python, C++, or Java), Machine Learning (ML) (scikit-learn), and Deep Learning (DL) frameworks (PyTorch or TensorFlow). Collects requirements and delivers Artificial Intelligence (AI) and ML solutions that drive customer and business value. Creates Web applications employing front-end technologies - React.js. Develops AI models on Cloud platforms - Amazon Web Services (AWS) Sagemaker. Uses visualization dashboard tools for result monitors - Tableau or Qlik Sense. Collaborates closely with the Product Owner to define tasks for upcoming sprints and manages stories, using Jira Tool. Oversees end-to-end AI/ML lifecycle management, including model versioning, data drift monitoring, and MLOps standards, using MLflow, Jenkins, or Amazon SageMaker Pipelines, to ensure scalable and reliable AI solutions in production.

Requirements

Bachelor's degree in Data Analytics, Computer Science, Engineering, Information Technology, Information Systems, Information Management, Business Administration, or a closely related field (or foreign education equivalent) and six (6) years of experience as a Director, Data Science (or closely related occupation) building algorithms using programming languages (Python, C++, Java, or Spark) and Machine Learning (ML) or Deep Learning (DL) frameworks (scikit-learn, Tensorflow, PyTorch, or Keras), to deploy applications in a financial services environment.
Or, alternatively, Master's degree in Data Analytics, Computer Science, Engineering, Information Technology, Information Systems, Information Management, Business Administration, or a closely related field (or foreign education equivalent) and four (4) years of experience as a Director, Data Science (or closely related occupation) building algorithms using programming languages (Python, C++, Java, or Spark) and Machine Learning (ML) or Deep Learning ( DL) frameworks (scikit-learn, Tensorflow, PyTorch, or Keras), to deploy applications in a financial services environment.
Demonstrated Expertise ("DE") performing advanced statistical modelling to develop, analyze, and evaluate supervised and unsupervised ML algorithms, using Neural Networks (RNNs (Recurrent Neural Networks), Seq-to-Seq, BERT (Bidirectional Encoder Representations from Transformers), Adversarial Networks, and LSTMs (Long Short-Term Memory)), Feature Selection, Clustering (Uniform Manifold Approximation and Projection (UMAP)), t-distributed Stochastic Neighbor Embedding(T-SNE), marketing attribution models, and treatment control matching using programming languages (Python, C++, or Java), within a financial services environment.
DE launching ML and DL models in online advertising (Clickstream data, Adobe, or Google analytics), Recommender Systems (Bandit algorithms, Bayesian models, NVIDIA Merlin, or Meta DRLM (Deep Learning Recommendation with Multi-Armed Bandits)), and user behavior applications (RNNs, BERT, LSTMs, or GANs (Generative Adversarial Networks)), using Python, C++, or Java to write production-level code and achieve greater performance; prototyping and deploying ML solutions using experimentation design (A/B Testing and Off-policy evaluation) within a financial services environment.
DE writing production-level code to deploy AI solutions, and achieve greater run-time performance and low latency according to a Test-Driven Development (TDD) mindset, using pytest (for unit and integration testing), Onnx Runtime, and Tensor RT (for optimizing model inference); automating build, test, and deployment of Docker-based ML models, using Jenkins and CI/CD pipelines; developing and deploying ML solutions and integrating caching mechanisms and client-server architecture, using Docker-containers in Cloud-based environments on AWS Sagemaker; building service endpoints using REST API, Flask, or Django; and developing front-end interfaces using ReactJS, within a financial services environment.
DE improving financial planning, advice offerings, and recommendations while liaising with business, product, and engineering stakeholder teams to assess the validity of ML models, using experimentation design; and communicating revenue or cost saving benefits to senior leadership within financial services environment, using business intelligence tools -- Seaborn, Altair, Streamlit, Plotly, Tableau, or Qlik.

Responsibilities

Develops and deploys AI models to address business needs by understanding the business problem, researching possible solutions, and prototyping AI capabilities.
Works closely with AI teams, business stakeholders, and deployment teams to ensure alignment with business objectives.
Trains and deploys advance DL and Natural Language Processing (NLP) models (RNNs, Seq-to-Seq, BERT, Adversarial Networks, LSTMs, GANs at scale.
Performs orchestration of training workflows, inference endpoints, and batch predictions.
Performs model evaluation, tuning, and scalability using distributed systems, parallel and multi-threaded programming techniques, and high-performance GPU environments.
Supports the operational deployment of AI/ML solutions.
Leads and oversees the full AI/ML lifecycle --data ingestion, model development, training, deployment, and monitoring.
Develops and delivers projects involving large-scale multi-dimensional databases and big data technologies, in collaboration with cross-functional teams and enterprise infrastructure.
Evaluates and makes decisions around the use of new or existing tools for a project.
Analyzes user needs and develops software solutions, applying principles and techniques of computer science, engineering, and mathematical analysis.
Researches, designs, and develops computer and network software or specialized utility programs.