Graph Data Scientist

Toronto, ON
Hybrid
Posted Nov 9, 2026
1 position

Opens LinkedIn

Employment type: Contract
Experience level: 2+ years
Minimum education: Bachelor’s degree
Posting language: English
Office presence: 3 days per week
Application method: Apply on LinkedIn (LinkedIn sign-in required)

Job summary

Design and develop graph machine learning models and knowledge graphs to identify relationships, communities, recommendations, risks, and anomalies in complex datasets. Build scalable data and model pipelines, help deploy solutions into production, and communicate findings and recommendations to technical and business stakeholders.

Job details

Graph Data Scientist 6-Month Contract with Potential For Extension and Conversion Hybrid - 3 days/week in office About the Opportunity We’re seeking a Graph Data Scientist to help develop advanced AI and machine learning solutions for a strategic client network project. This role focuses on uncovering meaningful patterns, relationships, communities, and predictive insights from large, highly connected datasets that traditional analytics methods may not fully capture. You’ll work at the intersection of graph machine learning, knowledge graphs, deep learning, and scalable AI, translating complex network data into actionable business intelligence and production-ready solutions. This is an opportunity to take ownership of graph data science initiatives, experiment with emerging techniques, and collaborate with data engineering, platform, product, and business stakeholders to bring innovative analytical solutions into production. Contract Details Duration: 6 months, with potential for extension and conversion Work arrangement: Hybrid, 3 days per week onsite Locations: Toronto, ON; Raleigh, NC; New Jersey; New York; or Jacksonville, FL Interview process: One virtual interview followed by one onsite interview What You'll Do Graph Machine Learning and Advanced Analytics Design and implement graph-based machine learning solutions, including node classification, link prediction, community detection, centrality analysis, similarity scoring, and anomaly detection. Develop graph embeddings and representation-learning models using techniques such as Node2Vec, GraphSAGE, graph neural networks (GNNs), and transformer-based approaches. Identify hidden relationships, influence patterns, communities, risks, and predictive signals across complex networks. Build and evaluate models for use cases such as recommendations, relationship discovery, fraud or risk detection, and network optimization, depending on business requirements. Apply graph theory and network analysis to translate ambiguous business challenges into measurable analytical outcomes. Knowledge Graphs and Data Engineering Design and enhance knowledge graphs by defining entities, relationships, ontologies, taxonomies, identity-resolution rules, and data provenance. Query and analyze graph data using Cypher, Gremlin, SPARQL, or equivalent graph query languages. Work with graph platforms such as Neo4j, TigerGraph, Amazon Neptune, or ArangoDB. Partner with data engineers to develop scalable data ingestion, entity-resolution, feature-engineering, and graph-processing pipelines using Python, SQL, Spark, Databricks, or similar technologies. Model Development and Production Deployment Build, train, tune, and evaluate deep learning models using PyTorch or TensorFlow, including graph neural networks and transformer architectures. Establish robust evaluation practices covering model performance, baseline comparisons, error analysis, explainability, bias, and business impact. Help transition prototypes and experiments into production through APIs, Docker, Kubernetes, CI/CD, model registries, and monitoring. Support reproducible ML workflows and production-ready solutions across AWS, Azure, or Google Cloud environments. Collaboration and Technical Communication Collaborate with engineering, product, platform, and business teams to move graph analytics from research and experimentation into practical applications. Present findings, visualizations, model limitations, and recommendations to technical and non-technical stakeholders. Document methodologies, assumptions, evaluation criteria, and model performance. Stay current with developments in graph AI, knowledge graphs, NLP, large language models, GraphRAG, and responsible AI. What We're Looking For Bachelor's or Master's degree in Computer Science, Data Science, Statistics, Applied Mathematics, Engineering, or a related quantitative discipline. A Ph.D. is an asset. At least 2 years of relevant professional experience in data science or machine learning, including meaningful hands-on experience with graph analytics, graph machine learning, or knowledge graphs. Strong understanding of graph theory and algorithms, including PageRank, shortest paths, connected components, centrality, community detection, similarity, and link prediction. Experience with at least one graph database or platform, such as Neo4j, TigerGraph, Amazon Neptune, or ArangoDB, and its associated query language. Advanced Python skills, including pandas, NumPy, scikit-learn, network analysis libraries, and sound software development practices. Hands-on experience developing and tuning deep learning models using PyTorch or TensorFlow. Familiarity with graph embeddings and representation learning, such as Node2Vec, Word2Vec, BERT-family models, or sentence embeddings. Strong SQL skills and experience working with large datasets using Apache Spark, Databricks, Hive, Presto, Trino, or comparable technologies. Experience deploying machine learning workloads in a cloud environment, ideally with Docker and Kubernetes. Strong analytical problem-solving skills, with the ability to validate assumptions, interpret model behavior, and turn complex findings into actionable recommendations. Excellent communication skills and experience working with cross-functional teams and business stakeholders. Nice to Have Production experience with graph neural networks and tools such as PyTorch Geometric, Deep Graph Library (DGL), Neo4j Graph Data Science, or TigerGraph Machine Learning Workbench. Experience designing enterprise knowledge graphs, ontologies, RDF/OWL models, semantic layers, SPARQL endpoints, or entity-resolution frameworks. Exposure to large language models, NLP, retrieval-augmented generation (RAG), or GraphRAG. Knowledge of recommendation systems, network optimization, advanced anomaly detection, reinforcement learning, or time-series forecasting. Familiarity with end-to-end MLOps, experiment tracking, feature management, model registries, automated testing, drift monitoring, and model retraining. Experience with Git, pull requests, CI/CD, infrastructure automation, and secure cloud engineering. Background in consulting or network-intensive industries such as financial services, telecommunications, cybersecurity, or healthcare. What Success Looks Like Developing accurate, explainable, and scalable graph models that address defined business needs. Improving the discovery of relationships, communities, recommendations, risks, or anomalies within complex datasets. Building reproducible and well-documented data and model pipelines. Helping stakeholders understand analytical findings and translating successful prototypes into production-ready solutions. Technology Environment Graph Databases and Knowledge Graphs: Neo4j, TigerGraph, Amazon Neptune, ArangoDB, Cypher, Gremlin, SPARQL Graph ML and Deep Learning: PyTorch, TensorFlow, PyTorch Geometric, DGL, GraphSAGE, Node2Vec, Transformers Data Science and Distributed Processing: Python, SQL, pandas, NumPy, scikit-learn, Apache Spark, Databricks, Hive, Presto, Trino Cloud and MLOps: AWS, Azure, GCP, Docker, Kubernetes, CI/CD, experiment tracking, model monitoring

What you’ll do

Requirements

Requires a bachelor's or master's degree in a quantitative discipline and at least two years of relevant data science or machine learning experience, including hands-on graph analytics, graph machine learning, or knowledge graphs. Candidates should have strong Python, SQL, graph theory, graph database, and deep learning skills, plus experience with large datasets, cloud deployment, and cross-functional communication.