The DCRS Inbound Correspondence Team is seeking a highly skilled Data Scientist to design, develop, and maintain document capture and intelligent processing solutions. This role combines machine learning, NLP, and emerging GenAI capabilities to transform large-scale structured and unstructured data into actionable insights and automated workflows.
The ideal candidate brings a strong foundation in software engineering, hands-on experience building enterprise-scale ML/NLP solutions, and exposure to modern open-source Gen AI models. This individual will operate within an Agile environment, partnering across technology and business teams to deliver impactful solutions.
Key Responsibilities
Design, build, and deploy machine learning and NLP models for document processing and correspondence analysis
Develop and maintain end-to-end ML pipelines, including:
o Data ingestion, cleansing, and transformation
o Feature engineering and model selection
o Performance tuning, evaluation, and visualization
Implement GenAI-driven solutions for document understanding (e.g., summarization, classification, entity extraction)
Build and optimize solutions leveraging OCR technologies for scanned and unstructured documents
Develop scalable systems to process large volumes of structured and unstructured data
Partner with cross-functional teams across Line of Business, Risk, and Technology to deliver solutions aligned with business needs
Contribute to software engineering best practices, including CI/CD, testing, and production deployment
Work within an Agile delivery model, supporting sprint planning, development, and release cycles
Required Qualifications
7+ years of experience as a Data Scientist or in a related role
Bachelor's degree in computer science, Data Science, or a related technical field
5+ years of programming experience, with strong proficiency in Python for ML/NLP development
Proven experience developing supervised and unsupervised ML models, including:
o Data analysis and feature engineering
o Model training, validation, and performance optimization
o Ensemble techniques and evaluation metrics
Hands-on experience building ML/NLP pipelines for large-scale datasets (structured & unstructured)
2+ years of experience developing enterprise ML/NLP solutions in one or more of the following:
o Named Entity Recognition (NER)
o Document Classification
o Document Summarization
o Topic Modeling
o Sentiment Analysis
o Dialog Systems
Experience working with OCR technologies and document processing solutions
Exposure to open-source GenAI models and frameworks (e.g., Hugging Face, LLaMA, Mistral, LangChain)
Experience working in Agile software development environments
Prefer red Qualifications
Master's degree in computer science, Data Science, or a related field
Strong experience collaborating with diverse stakeholders across multiple Lines of Business
Demonstrated ability to own and deliver initiatives independently
Strong analytical thinking and problem-solving capabilities
Experience navigating enterprise data ecosystems and large-scale platforms
Excellent communication and presentation skills for technical and non-technical audiences
Ability to prioritize and manage multiple initiatives in a fast-paced environment
Client-focused mindset with a strong emphasis on delivering business value