Verbal classification: Marek demonstrated professional ability in the field of artificial frustrated magnetism. He conducted a substantial amount of magnetic imaging experiments using magnetic force microscopy and run different numerical codes to extract relevant physical quantities. Marek is a serious, rigorous, hard working young scientist. We envision the publication of two articles in peer-review journals.

The master thesis of Bc. Marek Zalesak deals with the investigation of the artificial spin system based on the shakti geometry. This geometry was originally proposed by Chern et al. and further experimentally realized by Gilbert et al. Both authors showed how the Shakti spin configuration can be mapped onto a classic square spin ice, which then showed desired properties. The aim of this work was to revisit the experimental realization of this particular lattice and analyze the data in a more elaborate way, as was shown recently by Vojtech Schanilec. From the experimental point of view, the Gilbert’s results were reproduced although with a slightly less powerful demagnetization protocol. The deeper analysis of the data reveals that the underlying physics imprinted into the magnetic ordering is more complicated than previously reported by Gilbert and colleagues. The data were then analysed to identify all the vertex populations, which were then confronted with the Monte Carlo simulations. This was also confirmed by comparing the experimental and computed magnetic structure factors. The Monte Carlo model was then modified in order to reproduce the experimental data. In the final, but important, sections, it is concluded that also the mapping itself also brings some significant restrictions.
I personally find the work original and scientifically sound going beyond the groundbreaking work of Gilbert. The work is very mature containing all the key scientific ingredients: critical analysis of the literature, transparent data analysis and its critical evaluation. I am actually excited that one can reproduce the published data, which is something of great significance and which is rarely reported. It is also very transparent how the previous data were analysed and what the overlooked points were. The amount of experimental data and statistics is certainly sufficient to draw the well-supported conclusions. Mr. Zalesak also performed an extensive analysis of the Monte Carlo simulations, which were performed by the team at Neel Institute in Grenoble. The content of the manuscript is suitable and it contains all the necessary background to understand the experimental findings and the analysis. One weaker point is that sometimes it is slightly hard-to-follow some of the English expressions, which make some parts harder to understand. Nevertheless, the thesis as a whole is written at a very good level and I am looking forward to seeing this being published in a peer-reviewed journal.
I am certainly in favor to recommend this nice work to be defended in front of the examination committee. Topics for thesis defence:

1. 1. The whole idea stands on the mapping of spin configurations in shakti configuration to the square spin ice. Could you please in a pedagogical way explain step-by-step the whole idea of mapping as you would explain this to your younger colleague? Figures in Chapter 2.51 would be probably the best starting point. Also, what is the meaning of the final arrows of this mapping in the rightmost figures 2.6 and 2.7? How is the orientation of the arrows orientation defined?
2. 2. Following the first question, the whole idea of mapping relies on the fact that only Type C_I and C_II vertices are present. However, as shown in the thesis, this is not correct (This was puzzling me for the rest of the thesis and it is only discussed in the very last chapter, which is perhaps too late). Does it therefore mean that the whole analysis (here and in the work of Gilbert) does not really work?
3. 3. The proposed solution to the discrepancy between the experimental model and the Monte Carlo is connected to an intrinsic alternating magnetic field, which was implemented in the Monte Carlo code. However, the potential physical origin of this coupling is not discussed. Do you have some suggestions about the origin of this non-intuitive effective field?
4. 4. You discussed the difference between J_1 and J’_1 couplings. It would be straightforward to estimate the ratio by using micromagnetic simulations. Can you estimate this or justify your choice 𝐽′_1=0.9𝐽_1? I assume that there would be an interest to have these two couplings equal. Would it be achievable to adjust the coupling by, for example, reducing the width of the long magnet?
5. 5. The numbering of the bibliography items should be fixed in the final version.