Senior Data Scientist — Architect GenAI Platform (Hybrid)

Johnson Controls; São João da Madeira; Hybrid; EUR 70,000 - 95,000; Full time

Benefits offered by this job

Competitive salary and incentives

Greenfield platform architecture

Azure sponsorship and learning budget

Technical leadership path and mentored

Access to GenAI tooling and projects

Flexible hybrid working model

Job summary

Johnson Controls seeks a Senior Data Scientist to architect and ship a production-grade agentic AI platform that plans, reasons, and executes complex workflows. You will mentor engineers, lead GenAI initiatives, and collaborate across Azure cloud services.

The role demands hands-on experience with LLM-based applications, LangChain/LangGraph, and strong Python along with vectorDBs and SQL in enterprise environments. A hybrid work model is available.

Qualifications

• Proven experience building LLM/GenAI applications in production.
• Hands-on experience with agentic frameworks LangChain or LangGraph.
• Strong Python expertise (pandas, NumPy, scikit-learn) and API development.
• Experience with vector databases and embeddings (Azure AI Search, Pinecone, Weaviate, Qdrant, FAISS, or pgvector).
• Strong SQL skills and working with large-scale data.

Responsibilities

• Design, build, and productionize multi-agent systems including planning, tool calling, memory, and orchestration using LangGraph, LangChain, or similar frameworks.
• Develop end-to-end RAG pipelines: document ingestion, embeddings, and vector search, with grounding for agent knowledge.
• Integrate agents with enterprise systems and APIs via tool calling patterns and Model Context Protocol (MCP).
• Build and maintain LLM evaluation frameworks for agentic workflows and A/B testing.
• Implement guardrails, safety, and governance for agents: prompt-injection defense and audit logging.
• Fine-tune and optimize LLMs on Azure OpenAI / Azure AI Foundry (LoRA/PEFT, latency/cost trade-offs).
• Design, build, and evaluate classical ML models to complement agentic workflows.
• Own LLMOps/MLOps: experiment tracking, prompts, CI/CD, observability, drift monitoring, retraining strategies.
• Collaborate with data engineers on data quality and governance across Azure Data Lake, Databricks, and Synapse Analytics.
• Translate business problems into agentic AI solutions; present decisions to senior stakeholders.
• Mentor junior data scientists, lead code/design reviews, and promote engineering best practices.

Skills

LLM applications in production

LangChain

LangGraph

Python (Expert)

Vector databases & embeddings

SQL

Azure OpenAI

Tools

LangChain

LangGraph

Azure OpenAI

Pinecone

Weaviate

FAISS

pgvector

Azure Cognitive Search

Job description

Johnson Controls seeks a Senior Data Scientist to architect and ship a production-grade agentic AI platform that plans, reasons, and executes complex workflows. You will mentor engineers, lead GenAI initiatives, and collaborate across Azure cloud services.

The role demands hands-on experience with LLM-based applications, LangChain/LangGraph, and strong Python along with vectorDBs and SQL in enterprise environments. A hybrid work model is available.