Data Science Manager (Onshore )

Zensar•Charlotte, NC

4d•Hybrid

About The Position

Looking for a workplace where people realize their full potential, are recognized for the impact they make, and enjoy the company of the peers they work with? Welcome to Zensar! Read on for more details on the role and about us. What's this role about? We are looking for someone who can lead the teams through thought exercises and planning or roadmaps in addition to having some practical hands-on experience. Ideally someone who has transitioned from a hands-on data scientist to a strategic leader. Data Science & AI Manager - Healthcare Location: 3 days Hybrid (Charlotte, NC)

Requirements

Bachelor’s degree in a relevant field or equivalent professional experience.
6+ years of experience in data science, AI engineering, or applied ML, including 2+ years of team leadership or technical management.
Hands-on experience building agentic AI systems, including: Multi-agent workflows, Tool-using agents, Planning/monitoring agents
Strong experience with MCP servers or similar agent integration frameworks (e.g., LangChain tools, AutoGen, OpenAI tool calling).
Proficiency in Python, SQL, ML frameworks (PyTorch, TensorFlow, scikit-learn).
Experience with cloud data and compute platforms (Azure, Databricks, AWS, or GCP).
Strong understanding of LLMs, RAG pipelines, structured tool protocols, and knowledge graph integration.
Excellent communication, stakeholder partnership, and product-oriented thinking.

Nice To Haves

Experience with healthcare, foodservice, hospitality, or operational environments.
Familiarity with IoT data streams, workforce management systems, or real-time task operations.
Background in optimization, reinforcement learning, or continuous planning agents.

Responsibilities

Lead the strategy, architecture, and implementation of agentic AI systems for Healthcare Digital.
Design and manage MCP servers that provide structured, secure tool access for AI agents across platforms including meal ordering, food production, and EVS task management.
Build multi-agent systems with clear roles—e.g., planning agents, QA agents, data-retrieval agents, and operational copilots—that collaborate to support healthcare workflows.
Develop governance and routing layers that enable AI agents to safely execute tasks, call tools, generate recommendations, and interact with structured operational data.
Integrate agent-driven capabilities into Healthcare Digital’s platforms: Patient Meal Ordering: agentic nutrition checks, dietary rule enforcement, personalized recommendations. Food Production: prep-planning agents, demand forecasting agents, and waste-reduction optimization loops. EVS Task Management: task-ranking agents, routing agents, and real-time environmental monitoring copilots.
Build AI copilots for associates and managers that support decision-making, reduce administrative load, and automate repetitive tasks.
Ensure AI agents interact seamlessly with UI workflows, APIs, product logic, and underlying data systems.
Build and deploy predictive models that feed agent decision-making, including: Meal demand forecasting, EVS task prediction and prioritization, Labor and staffing optimization, Anomaly detection for operational issues
Integrate model outputs with MCP-based agents to create closed-loop automation—agents that both detect and act, not just analyze.
Translate findings into usable insights, dashboards, and operational recommendations for field teams.
Coach and mentor a team of data scientists, ML engineers, and AI engineers focused on agent development and MCP integration.
Partner with Healthcare Leadership (Culinary, EVS, Clinical Nutrition, Operations) to drive AI adoption and prioritize high-value opportunities.
Collaborate with IT, and enterprise AI teams to align on architecture, security, and platform standards.
Communicate complex AI and agent-based system concepts to non-technical stakeholders in clear, practical language.
Ensure all AI and agent systems adhere to governance frameworks, including privacy, compliance, and HIPAA.
Establish monitoring, auditability, and retraining workflows for both models and agents.
Implement agent safety controls, including sandboxed tool access, role-based permissions, and fallbacks for critical tasks.