Applied Data Scientist – Agentic AI Systems

About the role

Beamup builds an AI platform that turns supply chains into automated execution, using AI agents that detect issues, find root causes and run corrective actions across stores, distribution centers, warehouses and in-transit operations. It is looking for an experienced Applied Data Scientist in its R&D team who can design and build agentic systems and autonomous agents that turn supply chain solutions into a fully agentic platform.

Key responsibilities

• Design and build autonomous agentic systems that generate, configure and execute analytical pipelines to solve supply chain problems end-to-end
• Architect multi-agent workflows with planning, tool use, memory and feedback loops so agents can reason, adapt and improve
• Develop and integrate ML and deep learning models (predictive models, anomaly detection, demand forecasting) into agentic pipelines
• Research and apply state-of-the-art techniques in agentic AI, LLM orchestration and multi-agent systems to production use cases
• Translate logistics and supply chain challenges into agent-based problem formulations with product and domain experts
• Define evaluation frameworks for agent performance: correctness, reliability, robustness and edge-case handling
• Work with software engineers to productionize solutions, including testing, monitoring, versioning and iteration
• Contribute to team practices: reproducible code, experiment tracking, documentation and knowledge sharing

Required profile

• 4+ years of applied data science or ML in a product environment, with experience building agentic systems or autonomous agents
• MSc in Computer Science, Data Science, Mathematics, Statistics, Engineering or a related field, or equivalent practical experience
• Proven track record of multi-step agentic pipelines, including LLM-based agents, tool use, planning loops and memory mechanisms
• Experience evaluating models rigorously (baselines, cross-validation, error analysis) and defining evaluation strategies for agent behavior
• Strong communication and collaboration skills across engineering, product and supply chain experts

Required skills

• Hands-on experience with agentic frameworks such as LangChain, LangGraph or AutoGen
• Strong Python; familiarity with Spark for large-scale distributed data processing
• LLM API experience (OpenAI, Anthropic, Bedrock, open-source models) and prompt engineering for agentic use cases
• Nice to have: multi-agent architecture and orchestration at scale, cloud platforms (AWS, GCP, Azure) and MLOps, Docker and Kubernetes