This study draws on data from a three-year investigation involving third-year undergraduate students at Sofia University to examine the benefits of practice-based programming problems for school informatics education. The School Course in Informatics was part of the academic discipline that students in the “Pedagogy of Teaching Mathematics and Informatics” program studied. The training of pre-service teachers focused on creating tasks to test high school students' comprehension of branching and loop algorithms through both console and graphical interfaces.

Each task consisted of a problem description with input/output specifications and constraints and test cases along with assessment criteria and methodological hints which mirrored grading systems found in international competitions. This uniform structure facilitated clarity and compatibility with automated grading.
Three main problems emerged when pre-service teachers designed tasks according to research findings: ambiguous task boundaries together with vague data needs and unclear assessment standards. The majority of students struggled to distinguish major algorithmic mistakes from small syntax errors because these issues proved vital for formative assessment.

The research implemented metacognition together with reflection as core research elements. Students created two separate task sets which focused on branching and looping. Students received detailed instructor feedback for their first submission followed by participation in peer discussions. The assessment of the second assignment demonstrated students performed better due to the structured practice which enhanced their task quality alongside their teaching abilities.

The research demonstrates that building successful learning tasks stands as a fundamental competency for informatics instructors. The study identifies programming assignment quality into two categories which range from well-designed tasks that enhance learning through understanding evaluation and automation to poorly designed tasks which create obstacles to learning and effective feedback. The inadequate tasks generate a teaching-method assessment approach divide which reduces the effectiveness of instructional guidance.

Student-developed task piloting took place at Sofia High School of Mathematics with 8th, 10th and 11th grade students to test the research approach. The programming assignments followed the Bulgarian 8th-grade informatics curriculum—Topic 3: “Programming”, specifically subtopics 3.3 and 3.4 which dealt with branching and iteration. The assignments supported competency-based learning and curriculum integration.

Multiple-subject integration along with clearer instructions that match educational objectives emerges from systematic task development according to the study. The approach applies to all STEM fields because problem formulation and feedback prove essential regardless of field specificity.

The upcoming study will establish an open database with high-quality programming tasks to assist teachers in developing lesson plans and adjusting tasks while promoting teaching consistency. Well-constructed assignments support the integration of assessment and instruction while offering personalized feedback which stands as the foundation for superior informatics education.

The article is structured as follows:
1) Introduction,
2) Methodology,
3) Data collection,
4) Data analysis and Interpretation, and
5) Conclusions.

Keywords: Structured programming tasks, informatics education, task design, pre-service teachers, student assessment, grading criteria, STEM, teacher training, national curriculum, Bulgaria.