AI-based coaching for women in STEM subjects to improve academic success
Try out our metis platform?
The KI_FEM project
Our goal is to improve the motivation and academic success of women in STEM subjects through AI-based coaching. Due to the low proportion of women and high drop-out rates, there is an acute need for action.
Our solution metis is an AI coach as a personal assistant, outside of an e-learning platform and independent of a specific subject area or subject field. The AI coach supports female students in their intrinsic motivation, self-reflection on their own learning model and the development of individual learning strategies.
Aspects that metis can promote:
- Promote individual skills development and potential
- support self-directed and personalized learning
- reflect on your own self-assessment and learning model
- maintain or even increase personal motivation for the chosen field of study
Topics covered by metis:
- Learning strategies
- resilience
- self-organization
- self-efficacy
- stress management
- Scientific work
- Time management
Important documents for participants
Data protection of the funding body
Data protection of the project
Any questions, comments or suggestions?
Latest news
January 2026
Our major evaluation is coming to an end. How is Kira being received by students? What wishes and suggestions for improvement have emerged? The next development sprint is now being created in the spirit of participatory development.
October 2025
New design - metis is one step ahead. Thanks to student support in our Studium Generale seminar.
Our innovation is Kira, a specific coaching bot. Whether it's a conversation about learning strategies or student needs, Kira advises using systemic methods.
June 2025
Through focus groups, seminars, lectures and design thinking workshops, we share our Evaulations stand of Metis with prospective students and students. What features? Attractive quiz? Flow of conversation? Topics and content?
Thanks to the expertise of our students, we find out exactly which challenges require support in STEM studies.
January 2025
The MVP of our chatbot is ready for testing! Thanks to the first feedback loop with our Studium Generale course "Participatory Methods for Teaching and Learning", the first variant was launched. With 4 learning styles, the knowledge topics resilience and self-efficacy, our coach metis is ready.
October 2024
The context analysis is finished, the result are our personas that reflect the students. Curious? Our personas are published here.
July 2024
We are currently in the context analysis phase, where we want to get to know our target group. By inviting students from as many different degree programs as possible for an interview, we gain very diverse insights into different learning strategies and environments.
Publications
2025
-
Exploring conversational agents for novice programmers: a scoping review, Dec 2025, In: Discover Artificial Intelligence. 5, 1, 271Electronic (full-text) versionResearch output: Contribution to journal > Review article
-
Expectations and Needs of Female STEM Students for Academic Chatbots, 31 Aug 2025, MuC '25: Proceedings of the Mensch und Computer 2025. p. 565 - 570, 6 p.Electronic (full-text) versionResearch output: Contribution to book/conference proceedings/anthology/report > Conference contribution
-
Bridging the Treatment Gap: A Novel LLM-Driven System for Scalable Initial Patient Assessments in Mental Healthcare, 26 Apr 2025, CHI EA 2025 - Extended Abstracts of the 2025 CHI Conference on Human Factors in Computing Systems. Association for Computing Machinery, 116Electronic (full-text) versionResearch output: Contribution to book/conference proceedings/anthology/report > Conference contribution
-
Designing Effective Feedback in Educational Programming Assistants: Recommendations, Analysis, and Practical Guidelines, 2025, Mensch und Computer 2025 - Workshopband. Gesellschaft für Informatik e.V., 6 p.Electronic (full-text) versionResearch output: Contribution to book/conference proceedings/anthology/report > Conference contribution
2024
-
Effectiveness and Acceptance of Conversational Agent-Based Psychotherapy for Depression and Anxiety Treatment: Methodological Literature Review, 2024, Intelligent Systems and Applications: Proceedings of the 2024 Intelligent Systems Conference (IntelliSys) Volume 1. Arai, K. (ed.).p. 188–203, 16 p.Electronic (full-text) versionResearch output: Contribution to book/conference proceedings/anthology/report > Conference contribution
Contact and contact persons
Our team
© Michael Kretzschmar
Leitung Nachwuchsforschungsgruppe
NameDr.-Ing. Claudia Loitsch
Bedarfsorientiertes KI-Coaching für Studierende, Projektleitung
Send encrypted email via the SecureMail portal (for TUD external users only).
© M
Projektmitarbeiterin
NameMs Chrakhan Barzanji
Send encrypted email via the SecureMail portal (for TUD external users only).
Projektmitarbeiterin
NameMs Emilie Peters
Send encrypted email via the SecureMail portal (for TUD external users only).
© Sven Ellger/TUD
Projektmitarbeiter
NameMr Dipl.-Inf. Niclas Rosteck
Send encrypted email via the SecureMail portal (for TUD external users only).
Projektmitarbeiter
NameMr Dipl.-Inf. Sebastian Eberl
Send encrypted email via the SecureMail portal (for TUD external users only).
Projektmitarbeiter
NameMr Dipl.-Inf. Sebastian Rottmann
Send encrypted email via the SecureMail portal (for TUD external users only).
© Lucas Vogel
Research Associate
NameMr Julian Striegl M.Sc.
Send encrypted email via the SecureMail portal (for TUD external users only).
Visitor's Address:
Andreas-Pfitzmann-Bau, 1st floor, room 1057 Nöthnitzer Str. 46
01187 Dresden
None
Secretary on Campus Navigator
Student employees:
Sarah Turner
Emilie Peters
ScaDS.AI is home to the junior research group on AI-based coaching for students.
The SCS offers students a first point of contact for all questions relating to their studies and refers them to other advisory and expert services.