Cultural intelligence (CQ) framework as AI engineering primitives - ON-1222

Project type: Research
Desired discipline(s): Computer science, Mathematical Sciences, Cultural studies, Social Sciences & Humanities, Psychology
Company: Tulong Technologies Inc.
Project Length: 4 to 6 months
Preferred start date: 09/01/2026
Language requirement: English
Location(s): Toronto, ON, Canada
No. of positions: 1
Desired education level: Master'sPhDPostdoctoral fellowRecent graduate
Open to applicants registered at an institution outside of Canada: No

About the company: 

Tulong Technologies is an AI company building a plug-and-play cultural intelligence engine for enterprises and public-sector organizations. Tulong helps teams reduce and prevent AI bias by combining governed multicultural data with structured Cultural Intelligence (CQ) reasoning, confidence-aware scoring and escalation and human-in-the-loop expert review. Designed to integrate into existing AI workflows, the platform provides clear explanations, policy controls and audit-ready logs to support safer multilingual and multicultural communications and decision support across functions and industries.

Describe the project.: 

The primary goal of this project is to operationalize the academically validated Cultural Intelligence four-factor model (Earley and Ang, 2003) as functional, measurable AI engineering components within Tulong's Cultural Intelligence Engine (CIE).
The CQ framework comprises four factors: Metacognitive CQ (awareness of cultural assumptions), Cognitive CQ (cultural knowledge retrieval), Motivational CQ (confidence calibration) and Behavioural CQ (culturally-appropriate output adaptation). While this framework is well-established in organizational psychology, no prior research has translated it into executable AI engineering components. This gap represents both a novel research opportunity and a critical requirement for responsible cultural AI development.
The intern will design and implement each CQ factor as a discrete software module within the CIE's reasoning pipeline. The Metacognitive module will detect cultural assumptions in AI outputs using prompt engineering and fine-tuned classification. The Cognitive module will integrate Hofstede, GLOBE, Schwartz and Trompenaars cross-cultural frameworks into a retrieval-augmented generation system. The Motivational module will implement AI confidence calibration. The Behavioural module will produce culturally-adapted output recommendations using targeted language generation.
The intern will develop benchmark performance metrics for each module and conduct a comparative study measuring the CQ-enhanced pipeline against baseline LLM cultural performance across 10+ cultural communities. Deliverables include working software modules integrated into the CIE, a performance benchmark report and an academic paper establishing the methodology for translating social science constructs into AI engineering primitives.
This is novel research. No existing literature addresses CQ-to-AI translation at the component engineering level. The outcomes directly enable Layer 2 of the CIE, the reasoning engine on which all cultural bias detection and adaptation depends.

Required expertise/skills: 

Required: Large Language Model (LLM) prompting, fine-tuning and evaluation; retrieval-augmented generation (RAG) system design; Python programming (LangChain, LangGraph, HuggingFace Transformers); familiarity with Cultural Intelligence (CQ) theory (Earley and Ang, 2003; Van Dyne et al.); cross-cultural psychology frameworks (Hofstede, GLOBE, Schwartz).
Preferred: Organizational psychology background; intercultural communication theory; NLP benchmarking methodologies; AI fairness and bias detection research experience.
Software: Python (PyTorch, HuggingFace, LangChain), Jupyter notebooks, Git/GitHub, cloud compute (AWS/GCP), API frameworks for multi-model evaluation.
Assets: Prior experience with AI fairness research; exposure to multicultural marketing or HR contexts; multilingual capability is a strong asset for cross-cultural model evaluation.