Senior Scientist - Data Science & AI, Industrial Process Analytics

ppAre you excited to apply modern data science to real production plants, help process engineers act faster, smarter, and deliver measurable sustainability impact? Join our Industrial Process Analytics team to build advanced analytics that optimize chemical and biochemical manufacturing at scale. /ph3Key responsibilities /h3ulliDesign, build, and ship scalable data science applications for manufacturing - from first idea to validated, deployed product. /liliImplement model- and data-driven approaches that support plant engineers and operators in day-to-day decisions. /liliCollaborate in multi-disciplinary teams (data scientists, process/chemical engineers, cloud data engineers, reliability and production experts) to improve yield, throughput, quality, reliability, and sustainability. /liliGenerate insights and convert them into measurable impact. /liliChampion a data-driven mindset across sites; communicate results with clarity to technical and non-technical stakeholders. /liliContribute to best practices for data analysis, coding and MLOps. /liliStay up to date with state-of-the‑art methods, scout and implement new technology in‑house. /li /ulh3We offer /h3ulliDirect impact on manufacturing operations and sustainability KPIs. /liliA global Advanced Analytics team focused on optimizing biotech and chemical processes—fully aligned with dsm‑firmenich’s manufacturing excellence strategy. /liliReal‑world challenges and exposure to all businesses, with the opportunity to contribute strongly to both manufacturing operations as well as SR up‑ and downscaling. /liliAn experienced, supportive community that has optimized plants worldwide—and room to grow your expertise in both ML and manufacturing. /li /ulh3You bring /h3ulliPhD or similar experience in (Bio)Chemical Engineering or a related field, or Data Science/Statistics/Computer Science. /lili3-7 years of additional academic or industrial work experience in manufacturing data science, ideally for (bio)chemical processes. /liliStrong hands‑on Machine Learning expertise and rigorous, real‑world validation. /liliModeling for industrial settings: you can dive into methods and details, yet make pragmatic calls in a results‑driven environment. /liliExcellent communication and domain translation skills—partnering with engineers, operators, and leadership; navigating complex stakeholder landscapes. /liliExcellent problem‑solving skills and proven ability to work both independently and collaboratively. /liliAbility to create end‑to‑end computational workflows, from data ingestion to deployment and monitoring. /liliUnderstanding of (bio)chemical processes and process control, ideally including experience processing and modeling time‑series, tabular, and panel/longitudinal/multi‑way data; exposure to multivariate process analytics/chemometrics. /liliExperience with MPC or system identification is a strong advantage. /liliPhysics‑based and hybrid modeling (e.g., gray‑box, surrogate models, PINNs, digital twins) is a strong plus. /liliFamiliarity with vision and text‑based GenAI (for operator guidance, documentation mining, inspection, etc.) is a plus. /li /ulh3Technical skills /h3h3Advanced Analytics and Machine Learning /h3ulliPython and core Data Science stack for data manipulation, visualization, statistics, ML/DL, and time‑series/forecasting. /liliMultivariate modeling / Chemometrics for process monitoring and root‑cause analysis. /liliModel interpretability and uncertainty. /li /ulh3Software engineering lifecycle /h3ulliSoftware engineering best practices: git, code review, linting/formatting, unit/integration tests (pytest), packaging (uv), containers (Docker), exposure to CI/CD. /liliFamiliarity with data engineering and model management (DBT, databases, MLFlow). /li /ulh3Nice to have /h3ulliProcess analytics with Seeq or TrendMiner. /liliCausal robust modeling: DoE/experiment design, Bayesian methods, causal inference, drift detection. /liliHybrid control‑aware modeling: physics‑informed/gray‑box models, surrogates for optimization, MPC integration. /liliGenAI, LLMOps, agentic AI, Vision Language Models. /liliCloud platforms, AWS, Azure, Databricks. /liliOnline learning, IoT and edge scenarios, streaming and real‑time. /liliWorkflows: Nextflow/CWL or alternatives for reproducible pipelines, Cora cloud pipeline. /liliAPI and webapp development (FastAPI, Flask, Django, Streamlit, JavaScript). /li /ul /p #J-18808-Ljbffr