Pop-science.
You're eating the wrong sandwich, Uncle Fyodor. You do popular science for kids, but you need to come in from the side.
Employment of graduates. The agenda puts the emphasis on the student, but it should be put on the employer: develop the skills for a highly educated graduate, and not try to push in the unsuitable. The hands of those career centers are not enough for their thousands of students - the required number and frequency of cross-pollinating touches can only be built through the extensive involvement of employers who are not particularly prepared, and there are far fewer of them as counterparties to the interaction than students.
Student design bureaus. Supporting 15 universities with grants for something like this is a completely empty and even toxic story. The participation of students in the design movement lives in a lack of interest not of students, but of demand for it and skills in this industry. 15 artisan grants, targeted at 1,000 industrialists, would provide more support to the remaining hundreds of SKBs than just 15 independent projects per 1,000 students.
Popularization of science. Trying to sell a science career to students is a good thing. But the main target audience for popularization is not them. How to sell owners/managers of money for deposits of their use in science, technology and related projects - this is my personal popularization question, which seems to be of utmost importance right now.
Compete with conventional courier work in the mind of a young biologist or compete with conventional construction in the mind of a person with access to money. I personally favor the second.