Senior Data Scientist

Job Description:

Fidelity’s Enterprise Technology (ET) Artificial Intelligence (AI) Innovation group is seeking an experienced, highly motivated data scientist to research and build complex and strategic AI algorithms, models, platforms, and related technologies that will serve multiple Business Units (BUs) within the company. This candidate must be comfortable operating autonomously in a fast-paced environment and facing off with stakeholders at various organizational altitudes.

Over-arching responsibilities include assimilating requirements and delivering high value Artificial Intelligence/Machine Learning (AI/ML) solutions to drive customer and business value – from model evaluation, tuning and performance to multi-BU operationalization and model monitoring. Must have experience developing transformer-based models, e.g., BERT, RobERTa, DistilBERT, and is well-versed fine tuning open sourced generative LLMs, including adapter methods, e.g., LoRA, as well as advanced information retrieval approaches, e.g., Retrieval Augmented Generation/DPR/MMR and related methods.

Candidate must have a solid mathematical and modeling foundation developing supervised and unsupervised Machine Learning algorithms, e.g., regressions, decision trees, random forest, neural networks, feature selection/reduction, clustering, and parameter tuning. Performs validation and testing of models to ensure adequacy and reformulates models as necessary. The successful candidate will work collaboratively with distributed teams to prototype, learn, and quickly deliver data science solutions.

The Team

ET is the central and foundational technology division that serves all Fidelity businesses, clients, customers, and reputation by offering end-to-end IT services ranging from infrastructure, cloud, security, application development, as well as higher order reusable AI/ML fabrics, endpoints and pipelines spanning a variety of businesses.

Required Skills

• Master’s Degree or PhD in one or more of the following disciplines: Computer Science, Data Science, Mathematics, Computational Statistics, Physics, Machine Learning or STEM-releated areas with 6+ years proven experience.
• Experience curating massive amounts of data from multiple data sources, building predictive machine learning models and operationalizing these models.
• Experience with the complete range of AI/ML/DS models, e.g., supervised, semi-supervised and generative AI/ML algorithms.
• Knowledge of advanced statistical techniques and concepts (regression, properties of distributions, statistical tests and proper usage) and experience with applications.
• Strong programming, data science and high-performance compute environment skills, e.g., Python, R, SQL, Bash/Shell, CLI, PyTorch, SageMaker, Kubernetes/EKS, and CICD concepts, across on-prem and cloud platforms. CUDA experience a plus.
• Experience building solutions across databases in heterogeneous environment such as Snowflake, Postgres, and/or data streaming platforms.
• Consistent track record of understanding relationship between data and business objectives, developing mathematical models to predict business outcomes, and determining cause and effect relationships.

Other Desired Skills

• Self-starter, who can challenge status-quo, is curious and not afraid to ask “why?” and “what if?”, self-motivated to deliver high quality work in a timely manner.
• Ability to function in a fast-paced environment with multifaceted priorities, switching tasks if needed, building solutions collaboratively with others across the organization.
• Strong analytical and problem resolution skills, be able to do research, find answers to technical challenges, and learn new techniques.
• Ability to translate sophisticated topics to non-technical audiences.

The Value You Deliver

• Capturing and sharing new AI learnings, trends, methodologies, and technologies.
• Understand and articulate analytic requirements and outline high level integration designs to meet those requirements.
• Partner with team members and stakeholders to design and build a variety of reusable AI/ML packages and real-time models optimizing customer experiences with minimal friction and achieving measurable outcomes.
• Build systems and deploy models that can assemble and compute real signals within milliseconds and at an event level.
• Identify and integrate new data sources to continuously improve models.
• Develop model testing frameworks for performance and quality. Prepare model governance documentation.
• Develop processes and tools to monitor and analyze model performance and making vital adjustments.

Certifications:

Category:

Fidelity’s hybrid working model blends the best of both onsite and offsite work experiences. Working onsite is important for our business strategy and our culture. We also value the benefits that working offsite offers associates. Most hybrid roles require associates to work onsite every other week (all business days, M-F) in a Fidelity office.