About the role
AI summarisedThis is a mid-level data science role at STMicroelectronics, a global semiconductor company. The position focuses on designing and building agentic AI-driven expert systems for predictive and autonomous decision-making in manufacturing, involving multiagent AI, simulation, digital twins, and integration of predictive models.
IDMFull-timeData Science
Key Responsibilities
- Design and build agentic AI–driven expert systems that enable predictive and autonomous decision making in manufacturing.
- Develop agent based and multiagent AI systems that can plan, decide, and act on behalf of operators, equipment, or production lines.
- Design reasoning and decision policies that leverage rules, optimization, and learning to handle complex, real time manufacturing scenarios (e.g. dispatching, scheduling, routing, quality control).
- Use simulation and digital twins to model production flows, equipment behavior, and system constraints, and to safely test and refine agent strategies before deployment.
- Integrate predictive models (e.g. for demand, machine health, cycle time, yield) into expert systems to support proactive and prescriptive decisions.
- Collaborate closely with process owners, planners, and equipment engineers to translate operational knowledge into rules, heuristics, and agent objectives.
- Evaluate agentic AI solutions using KPIs such as throughput, WIP, cycle time, OEE, and cost, and iteratively improve performance.
- Document architecture, algorithms, experiments, and support presentations, publications, and IP creation related to expert and agentic AI systems.
Requirements
- Bachelor or master's in computer science, AI, Data Science, Operations Research, Industrial/Electrical Engineering, or similar.
- Strong foundation in AI and machine learning, with exposure to at least one of: Reinforcement learning / multiagent RL, Planning and decision making under uncertainty, Rule based or knowledge-based systems.
- Experience with Python and AI/ML ecosystems (e.g. PyTorch, TensorFlow, scikitlearn, RL libraries).
- Familiarity with simulation and modeling of manufacturing or logistics systems (e.g. discrete event simulation, agent-based simulation, digital twins).
- Understanding of operations research / optimization (e.g. scheduling, routing, resource allocation) is a strong plus.
- Comfortable working with real operational and equipment data, including cleaning, feature engineering, and validation.
- Strong analytical and conceptual skills, able to formalize complex shopfloor logic into rules, constraints, and agent objectives.
- Collaborative, curious, and proactive, able to work with cross functional teams and incorporate domain expert feedback into AI systems.
- English business proficiency, both written and spoken, for documentation, stakeholder discussions, and technical presentations.