Data Scientist - Graph Neural Networks (GNN)

We are seeking a highly skilled Data Scientist with proven expertise in Graph Neural Networks (GNNs) and Graph Machine Learning to lead the design, development, and implementation of graph-based AI models as part of a strategic Proof of Concept (POC). The GNN architecture is the technical core of this engagement and, therefore, candidates must demonstrate prior hands-on experience building, training, evaluating, and deploying graph-based machine learning solutions.

General Data Science, Machine Learning, or Deep Learning experience alone will not be considered sufficient.

Key Responsibilities

• Design, build, and optimize Graph Neural Network (GNN) models for complex business problems.

• Develop graph-based solutions for:

• Link Prediction
• Node Classification
• Recommendation Systems
• Network Analysis
• Knowledge Graph Analytics
• Fraud Detection

• Entity Resolution

• Build scalable graph data pipelines and feature engineering workflows.
• Work with large-scale graph datasets and graph databases.
• Conduct model evaluation, experimentation, and performance optimization.
• Collaborate with domain experts, architects, and engineering teams to deliver production-ready solutions.
• Present technical findings and solution recommendations to stakeholders.

Proven hands-on experience implementing:

• Graph Convolution Networks (GCN)
• Graph Attention Networks (GAT)
• GraphSAGE
• Heterogeneous Graph Networks
• Temporal GNNs

Experience solving real-world Graph ML problems.

1. Demonstrated Graph ML Delivery Experience

Candidate must provide examples of prior graph-based machine learning implementations, including:

• Problem statement
• Graph modeling approach
• Architecture used
• Business outcome achieved

Note: Prior experience in power systems is not mandatory. However, prior Graph ML/GNN implementation experience is mandatory.

1. Python & Advanced Machine Learning

Strong experience with:

• Python
• NumPy
• Pandas
• Scikit-learn
• Data processing and feature engineering

1. GNN Frameworks

Hands-on expertise with:

• PyTorch Geometric (PyG)
• Deep Graph Library (DGL)
• TensorFlow GNN

1. Deep Learning

Experience with:

• PyTorch
• TensorFlow
• Neural network design
• Hyperparameter tuning
• Model optimization

1. Graph Data Modeling

Experience working with:

• Node and edge feature engineering
• Graph embeddings
• Knowledge graphs
• Graph representation learning

1. Communication & Stakeholder Management

Ability to explain complex graph-based concepts to business stakeholders.

Experience working in cross-functional delivery teams.

Should-Have Skills

Graph Ecosystem

• Neo4j
• TigerGraph
• Amazon Neptune
• JanusGraph
• NetworkX

Advanced AI & GenAI

• GraphRAG
• Knowledge Graphs
• LLM + Graph integrations
• Retrieval-Augmented Generation (RAG)
• LangChain / LangGraph

Cloud & MLOps

• Azure Machine Learning
• AWS SageMaker
• Google Vertex AI
• Docker
• Kubernetes
• CI/CD for ML

Big Data Technologies

• Spark
• GraphX
• Databricks
• Snowflake
• Delta Lake

Domain Experience (Preferred)

• Utilities
• Energy
• Power Systems
• Supply Chain
• Telecommunications
• Financial Services

Mandatory Qualification Criteria

• Demonstrated experience building Graph Neural Network (GNN) or Graph Machine Learning solutions.
• Must provide at least one detailed Graph ML implementation example.
• Strong hands-on knowledge of graph modeling, graph embeddings, and GNN architectures.