Title

Question-Time Reasoning over Knowledge Graphs using LLM-Augmented Retrieval

Objective

Design a system that takes natural language questions and retrieves relevant knowledge from a graph to support accurate, grounded answers by an LLM. This includes:

• Finding relevant subgraphs or embeddings from the KG.
• Integrating structured knowledge into LLM prompts or reasoning.
• Evaluating the benefit of using a graph versus traditional retrieval.

Scope & Challenges

• Translate natural language queries into graph-level operations (e.g., node/entity linking, subgraph retrieval, graph traversal).
• Use graph embeddings, Graph-RAG, or symbolic reasoning to integrate KG knowledge into answers.
• Ensure faithfulness, latency, and usefulness in an LLM-based QA setting.
• System must consume graphs from GrailQA (public) or other proposed source.

Dataset

• GrailQA benchmark:
• Subset of WikiData with real-world nodes and edges.
• 64k+ natural language questions.
• Gold answers and logical forms.
• Additional test queries may be defined against internal graphs we are building at Hypertype.

Deliverables

• A question-answering pipeline that:
• Retrieves relevant KG subgraphs based on a question.
• Integrates structured context into LLMs via prompting or hybrid approaches.
• Support for both real and benchmark KGs.
• Evaluation report with:
• QA accuracy (with vs. without KG).
• Faithfulness to graph facts.
• Latency benchmarks and ablation studies.

About Hypertype

We’re building the next generation of AI products and AI Agents in the Customer Support space. Focused on manufacturing but with clients across other sectors like fintech, health tech and more we are giving hundreds of companies across the globe the capacity to transform their customer journey with the latest advancements in AI. Our customers proudly recognise us for delivering the "highest quality answers in the market... outsmarting Gemini, Fin, Microsoft Copilot etc". —a testament to our unwavering commitment to excellence and precision in every interaction.

We are a group of people that work the extra mile together, that fearlessly push the boundaries to new heights and that deeply own what they deliver. We look for those who are ambitious, humble and fast to build together things humankind has not witnessed yet.

Department

Hyperbrain AI

Locations

Stockholm