RESEARCH

Data-driven education

My work connects research with hands-on system development. I study how people learn with technology, and I build software that supports programming education, automated feedback, intelligent assistance, and data-informed teaching.

Programming skills acquisition

I develop and study learning environments that help students build programming skills through practice, feedback, and interaction data.

Automating feedback

I design feedback systems that help learners receive timely, actionable support on programming tasks.

Large language models

I explore how large language models can support teaching, feedback, and educational research in realistic learning contexts.

JeepyTA - virtual teaching assistant

JeepyTA is an example of my research-through-building approach: an AI-powered teaching assistant designed to support students through contextual help, adaptive guidance, and scalable educational interaction.

To play a wrong note is insignificant; to play without passion is inexcusable
— Ludwig van Beethoven

After hours

Outside research and software development, I enjoy running, playing soccer, and creating music.

Guitar has been part of my life since childhood, and over the years I have recorded original songs and collaborated with other musicians in live performance. These creative interests shape how I think about design, expression, and building things that connect with people.

guitar.pankiewicz.eu

Chopin Died is a band I founded. We released several albums and performed at numerous concerts. If you'd like to learn more, visit:

chopindied.com
Left image

Selected talks

I regularly share work on AI in education, learning analytics, programming education, and intelligent educational systems in conference presentations, invited talks, and workshops.

  • From ChatGPT to Classroom Impact: Causal Evidence for LLMs in Education

    Keynote talk at The Learning Ideas Conference 2026
    New York, NY, USA
    2026-06-12

  • Generative AI in Education: Enhancing Learning, Feedback, and Research with LLMs

    Keynote talk at Workshop AI for Education - Tools, Opportunities, and Risks in the Generative AI Era
    The 39th Annual AAAI Conference on Artificial Intelligence
    Philadelphia, PA, USA
    2025-03-03

  • JeepyTA: an LLM-Powered Virtual Teaching Assistant

    Presentation at the Digital Learning Lab
    Online, University of California, Irvine, CA, USA
    2025-01-31

  • Innovative Applications of Large Language Models in Research: A Focus on Education

    Presentation at the PAN-Metrics - Seminar on Novel Measurement Problems in Social Sciences
    Online, Polish Academy of Sciences, Poland
    2024-11-18

  • Students’ Use of an LLM-Powered Virtual Teaching Assistant for Recommending Educational Applications of Games

    Presentation at the 10th Joint International Conference on Serious Games - JCSG 2024
    New York University, New York, NY, USA
    2024-11-07

  • Ordered network analysis in CS education: unveiling patterns of success and struggle in automated programming assessment

    Presentation at the 29th annual ACM conference on Innovation and Technology in Computer Science Education (ITiCSE 2024)
    Università degli Studi di Milano, Milan, Italy
    2024-07-09

  • Navigating Compiler Errors with AI Assistance - A Study of GPT Hints in an Introductory Programming Course

    Presentation at the 29th annual ACM conference on Innovation and Technology in Computer Science Education (ITiCSE 2024)
    Università degli Studi di Milano, Milan, Italy
    2024-07-08

  • Comparison of Three Programming Error Measures for Explaining Variability in CS1 Grades

    Presentation at the 29th annual ACM conference on Innovation and Technology in Computer Science Education (ITiCSE 2024)
    Università degli Studi di Milano, Milan, Italy
    2024-07-08

  • Large Language Models and Learning Analytics

    Tutorial at the SoLAR Learning Analytics Summer Institute
    University of Texas at Arlington, Arlington, TX, USA
    2024-06-11