Senior Scientist, Dynamic Context Graphs & AI Agents

Accrete; Boston (MA); On-site; USD 160,000 - 210,000; Full time

Job summary

Accrete is seeking a Senior Applied Scientist to advance its Knowledge Engine through context graphs and agentic systems. You will work on knowledge representation, graph-based ML, and LLM-driven extraction at scale.

You will design dynamic graphs, evaluate reasoning with provenance and uncertainty, and partner with engineers to turn research into production capabilities. This role emphasizes rigor, publication potential, and cross-team collaboration.

Qualifications

• Ph.D. in a related technical field with ML/NLP/graph knowledge.
• Strong track record in knowledge representation, graph ML, NLP, or IR.
• Publication record in NeurIPS/ICML/ACL/EMNLP/KDD/WWW/ISWC.
• Experience building extraction pipelines over unstructured and heterogeneous sources, including LLM-based extraction.
• Experience designing evaluation for problems without clean ground truth.
• Strong Python and PyTorch skills; experience with Neo4j and graph learning frameworks.
• Ability to frame research questions independently and communicate technical work across teams.

Responsibilities

• Conduct applied research focused on context graphs, knowledge representation, graph intelligence, and AI systems operating over complex data.
• Develop algorithms for constructing and updating dynamic context graphs from structured and unstructured data.
• Research entity, relationship, event extraction, and temporal reasoning across sources.
• Design graph representations of entities, relationships, events, and organizational knowledge.
• Incorporate time, provenance, uncertainty, and source attribution into graph models.
• Develop graph-based methods to support LLM reasoning, retrieval, and agentic workflows.
• Explore integration of LLMs with graph representations for grounding and explainability.
• Prototype graph algorithms and collaborate with engineering to scale production.
• Create benchmarks and evaluation frameworks for graph quality and retrieval performance.

Skills

Python

PyTorch

Knowledge graphs

NLP

Information retrieval

Graph learning

Research methodology

Education

PhD in CS/EE/CL/CogSci

Tools

Neo4j

PyTorch Geometric

DGL

Job description

Accrete is seeking a Senior Applied Scientist to advance its Knowledge Engine through context graphs and agentic systems. You will work on knowledge representation, graph-based ML, and LLM-driven extraction at scale.

You will design dynamic graphs, evaluate reasoning with provenance and uncertainty, and partner with engineers to turn research into production capabilities. This role emphasizes rigor, publication potential, and cross-team collaboration.