Senior Data Scientist (Agentic AI Platform)

Description

The company provides thermal management, mission-critical building systems, energy efficiency and decarbonization, helping customers use energy more productively and cut carbon emissions.

Its customers include operators in data centers, healthcare, pharmaceuticals, advanced manufacturing and higher education. The company has been operating for more than 140 years.

The team is building an agentic AI platform: autonomous, tool-using agents that plan, reason and carry out multi-step business workflows from start to finish. This senior data scientist role is central to that work and to setting the technical direction for GenAI projects across the organization.

Responsibilities

• Lead work on a platform of autonomous agents that use tools to plan, reason and run business workflows end to end, drawing on strong ML foundations and hands-on LLM application skills.
• Build and ship production agentic systems covering planning, tool and function calling, memory and orchestration, using LangGraph, AutoGen, CrewAI or Semantic Kernel.
• Create RAG pipelines: ingest documents, choose chunking methods, generate embeddings, run vector search with Azure AI Search, FAISS or pgvector, re-rank results and ground agent knowledge.
• Link agents to internal systems and APIs through tool calling, Model Context Protocol (MCP) or similar connectors.
• Build and maintain evaluation frameworks for agent workflows: task completion, hallucination detection, trajectory analysis, LLM-as-judge pipelines and A/B tests.
• Put guardrails in place: defenses against prompt injection, content filtering, role-based tool permissions, human-in-the-loop checkpoints and audit logs.
• Fine-tune and optimize LLMs where needed (LoRA/PEFT, prompt optimization, model routing, latency and cost trade-offs) on Azure OpenAI and Azure AI Foundry.
• Design, build and assess classical ML models (classification, regression, forecasting, NLP) where they support agent workflows.
• Own LLMOps and MLOps for the platform: experiment tracking, prompt versioning, CI/CD, observability and tracing with LangSmith, Azure Monitor and OpenTelemetry, drift monitoring and retraining.
• Partner with the data engineering group to keep data quality, availability and governance sound across Azure Data Lake, Databricks and Synapse Analytics.
• Turn unclear business problems into agentic AI solutions and present architecture decisions and results to senior stakeholders.
• Coach less experienced data scientists, lead reviews of code and design, and promote engineering best practices.

Requirements

• Professional experience in data science or ML, including LLM or GenAI applications running in production.
• Hands-on work with LangChain or LangGraph.
• Solid grasp of LLM application patterns, including prompt engineering.
• Expert Python: pandas, NumPy, scikit-learn, async programming and FastAPI API development.
• Experience with vector databases and embeddings, such as Azure AI Search, Pinecone, Weaviate, Qdrant, FAISS or pgvector.
• Strong SQL skills and experience working with large-scale data.

Conditions

Flexible hybrid working model.

Benefits

• Competitive salary plus performance-based incentives.
• Chance to design a greenfield agentic AI platform from the start.
• Sponsorship for Azure and AI certifications, plus a continuous learning budget.
• A technical leadership track and mentoring opportunities.
• Access to leading GenAI tools, compute and cross-domain AI projects.
• A collaborative culture.