A pedagogical model where human mentors and AI tools jointly support student learning in project-based contexts. Human mentors provide conceptual guidance, debugging, and problem formulation support; AI tools accelerate execution, code generation, and rapid iteration. Demonstrated by Chawla et al. (2026) in a financial forecasting project with high-school students.

Human-AI Co-Mentorship

Definition

A pedagogical model where human mentors and AI tools jointly support student learning in project-based contexts. Human mentors provide conceptual guidance, debugging, and problem formulation support; AI tools accelerate execution, code generation, and rapid iteration. Demonstrated by Chawla et al. (2026) in a financial forecasting project with high-school students.

Key Insight: Workflow Design over Traditional Instruction

Instead of "learn theory first, apply later," students identified the sequence of steps needed to solve their problem and used AI tools to execute each step. This just-in-time approach enabled students with limited backgrounds in AI and finance to build meaningful predictive models. AI tools served as scalable accelerators, not replacements for human mentorship.

Connections to Agentic Workflows

This model bridges Agentic Workflows Education and practical classroom implementation. Where agentic workflows describe AI agent paradigms (reflection, planning, tool use, multi-agent collaboration), co-mentorship shows how those paradigms manifest in human-AI educational partnerships. The daily stand-up structure echoes the reflection loop in Agentic Education Coding.

Connected Concepts

Connected Articles

Citation

Khvatskii, A.F.C.A.C.R.S.J.G.G. (2026). Human-AI Co-Mentorship in Project-Based Learning: A Case Study in Financial Forecasting