Senior Data Scientist

Working with Us

Challenging. Meaningful. Life-changing. Those aren't words that are usually associated with a job. But working at Bristol Myers Squibb is anything but usual. Here, uniquely interesting work happens every day, in every department. From optimizing a production line to the latest breakthroughs in cell therapy, this is work that transforms the lives of patients, and the careers of those who do it.

You'll get the chance to grow and thrive through opportunities uncommon in scale and scope, alongside high-achieving teams. Take your career farther than you thought possible.

Bristol Myers Squibb recognizes the importance of balance and flexibility in our work environment. We offer a wide variety of competitive benefits, services and programs that provide our employees with the resources to pursue their goals, both at work and in their personal lives. Read more:

Summary:

As a Senior Data Scientist within Bristol Myers Squibb's AI Venture Studio delivery team, you will be a hands-on senior individual contributor who helps convert ambiguous scientific and business opportunities into measurable AI product hypotheses, experiments, and working solutions. You will partner with AI Engineers, Data Engineers, App/Cloud Engineers, Frontend Engineers, product owners, and domain experts to build and evaluate AI systems across R&D, Commercialization, Manufacturing, and Enabling Functions.

The role sits at the edge of applied data science and AI engineering: you will design evaluation datasets, build analytical features, prototype models, test agent and retrieval performance, measure product impact, support sandboxed data problem solving, and explain model behavior in ways stakeholders can trust. You will help define the good analytical context agents need to perform reliable work, including query history, column values, explicit instructions, memory, data tools, warehouse context, and curated source meaning.

BMS is an AWS-first engineering environment for these products, and your work will use AWS-aligned data and AI services alongside enterprise-preferred tools such as Open Search, Amazon S3 Vectors, Amazon Neptune, Postgre

SQL/RDS, Lang Graph, Lang Smith, and a variety of approved frontier LLM models and APIs. This is a role for someone excited to work hands-on with the latest AI tools and frontier technologies, pushing the limits of what technology can do to help BMS discover, develop, and deliver innovative medicines.

Key Responsibilities:

AI/ML Experimentation and Product Prototyping:

* Frame ambiguous business and scientific questions into measurable AI product hypotheses, success metrics, evaluation plans, and rapid experiments.

* Contribute to six-sprint, 12-week AI Accelerator agile cycles by testing hypotheses, validating AI product increments, and adapting analyses during two-week sprints.

* Build data science prototypes using Python, SQL, notebooks, APIs, and AWS-aligned data services.

* Support sandboxed data problem solving in non-production environments, enabling agents and analysts to branch, transform, test, and audit code-plus-data experiments before promotion.

* Evaluate and curate the analytical context agents and analysts rely on, including explicit instructions, memory, data tools, and curated meaning from source materials and recommend improvements based on measured impact on agent quality. Develop analytical features, embeddings, classifiers, ranking/scoring methods, recommendation logic, simulation approaches, or optimization methods as needed for product outcomes.

* Partner with Data Engineers to shape reliable datasets, retrieval corpora, metadata, and feature pipelines using S3, Athena, Postgre

SQL/RDS, vector databases, and knowledge graphs.

Agentic AI, Retrieval, and Evaluation Science:

* Design and execute evaluations for LLM, RAG, and agentic workflows, with emphasis on context quality, knowledge curation, semantic evolution, and model quality.

* Build evaluation rubrics, golden datasets, structured output validation, error taxonomies, hallucination risk measurement, and SME review loops.

* Use tools such as Lang Graph, Lang Smith, Pydantic

AI, or similar frameworks to test agent behavior, retrieval quality, reasoning traces, and workflow reliability.

* Evaluate whether curated enterprise context improves agent quality, reliability, traceability, and decision usefulness compared with raw document retrieval.

* Assess model and agent outputs for quality, uncertainty, calibration, bias, hallucination risk, traceability, and fitness for intended use.

* Explore approved proprietary and open model options through enterprise channels and recommend model/task pairings based on evidence, risk, cost, and performance.

Decision Science, Analytics, and Impact Measurement:

* Define KPIs and analytical measurement plans for AI products, including adoption, user behavior, workflow efficiency, scientific utility, and business value.

* Use bi-weekly demos, sprint reviews, stakeholder feedback, and performance results to…