top of page


At CMU, I gained knowledge, skills and hands-on experience of how to apply the evidence-based approaches to inspect and fill the gaps between applied learning science research and the real-world education practice.


For educational curriculum, learning program or product design, the following design framework is my "big idea" that guides my design process: 






































































Note that I usually modify this design process case by case to meet the design goals. For example, time and budget are two important factors to affect the design process. The investment of time and money to the design should be paid off based on the driven evidence.  


This "big idea" also illustrates the skillsets that a learning engineer or learning product designer should be familiar with. I gained solid, hands-on training from METALS program to learn and practice this design process and relevant skills. Please see some example projects I have done at CMU and my reflections on them.

Binary 101 is an intelligent tutor designed to help lower grades students learn the binary concept. This is the workshop project of 2016 CMU LearnLab Summer School.

My role in this two people teamwork: set learning objectives for each lesson, defined and decompose the skills/ knowledge components, designed the activities, create the behavior graph, designed and coded the interface.

Please see the project details here.

Personalized E-Instruction of Exposure is a prototype of an intelligent tutor designed for adult learners to learn how to use manual setting of exposure of DSLR camera. The project prototypes how the e-instruction can be adaptive to learners' prior knowledge and personal interests. This is my independent course project of E-Learning Principles at CMU.

Please see the project details here.

Cubic Math Tutor is a prototype of an intelligent math tutor customized for two private elementary school in China. It allows math teachers to monitor and provide ,on-the-spot support for students during blended class sessions, and helps parents to understand their children's performances through the personalized analytics report. This is my independent study project under advice of Dr. Bruce Mclaren at CMU.

Please see the project details here.

MOOC Learner Behavior is a data analysis of Stanford MOOC Student Post Dataset. We visualized our analysis findings for our guiding question:  "what are the confusion characteristics of MOOC blog posts for different users over the course duration?".

My role in this five people teamwork: select the dataset, cleaned the dataset, defined the guiding question, designed the visualization, conducted the user testing. 

Please see the project details here.

Data analysis of stress level of college students is a course project that not only helps college students have knowledge of their own stress experience, but also understand the complexities of the experiences of their peers.  We chose the StudentLife Dataset, which contains data collected from 48 undergraduate and graduate students at Dartmouth, to find common student patterns, form the stress model, and test our model that associated with high levels of stress.  

My role in this five people teamwork: analyze the data, defined the research questions, designed the visualization, conducted the user testing. 

Please see the project details here.

bottom of page