BN31P2-Data Scientist

Description BN31P2- Data ScientistCAS Columbus, OH CAS uses unparalleled scientific content, specialized technology and unmatched human expertise to help R&D organizations across Commercial, Government and Academic sectors create groundbreaking innovations that benefit the world. As the Scientific Information Solutions Division of the American Chemical Society, CAS manages the largest curated reservoir of scientific knowledge, and for 119 years, has helped innovators mine, assess and apply that information to keep businesses thriving. The CAS team is global, diverse, endlessly curious and strives to make actionable scientific insights accessible to innovators worldwide. CAS is currently seeking a Data Scientist. This position will be located in our headquarters in Columbus, Ohio. Job Summary This is an individual contributor role on the Data Analytics and Insights (DAI) team, which builds the AI that powers CAS's scientific information products. As a data scientist, you will build the models, agents, and AI-powered features that ship in those products, working alongside data engineers, product managers, and scientific domain experts. This role requires dual-domain expertise: you must be both a capable software engineer and formally trained in a scientific discipline relevant to CAS's customers (chemistry, life sciences, materials science, or a related field). DAI builds AI that is grounded in CAS's scientific content rather than layered on top of it, so the people who build it need to read the science as well as write the code. You will work on features such as the Newton research assistants in SciFinder and BioFinder, the natural-language query agent in IPFinder, predictive models that ship inside products (for example property, toxicity, or biologic developability prediction), and AI-assisted content curation pipelines, using your scientific training to judge whether their outputs are correct and useful to working scientists. Success in this role requires strong programming fundamentals, formal scientific training, a genuine interest in agentic AI, and a collaborative approach to building reliable, production-grade systems. Candidates who bring only software engineering or only scientific training are not a fit for this position. Job Accountability Develop and deploy agent-based workflows, retrieval-augmented generation (RAG) pipelines, and LLM-powered features within CAS products using Python, LangGraph/LangChain, and related tools. Develop and deploy machine learning models that ship inside CAS products (for example property, toxicity, or biologic developability prediction), using your scientific training to inform feature design and to validate model behavior. Ground agentic and predictive features in CAS's authoritative scientific content, and use your scientific training to judge whether system outputs are accurate, defensible, and useful to researchers. Engineer for production, not for demos: deliver clean, tested, well-documented software with appropriate safeguards such as content safety guardrails, entitlement-based access control, evaluation, and model fallback, following team engineering standards (unit, integration, and end-to-end tests, containerized development, and CI/CD practices). Work with product managers, scientific domain experts, data engineers, and other teams to translate scientific and research requirements into working technical solutions. Contribute to evaluation frameworks for GenAI systems, focusing on accuracy, reliability, scientific correctness, and continuous improvement of agentic features. Build and operate solutions on cloud infrastructure (AWS, Azure, GCP, or similar), supporting the production deployment and ongoing operations of AI systems. Use agentic software development tools such as Claude Code as a core part of