In astronomy, precise determination of stellar positions, proper motions, and parallaxes based on space telescope observations is a Big Data problem. It requires a dedicated software solver running on a high-performance computer to analyse billions of input data records and produce an output stellar catalogue. The solution process relies on a sophisticated model to calibrate out the distortions, which are inevitably presented in the raw input data due to the imperfections of the telescope. After the solution is calculated, its quality must be assessed for physical correctness, scientific value, and possible ways of calibration model improvement. The tools for the solution quality assessment are as important as the solver itself and contribute to the solver’s tractability by unveiling the path to fine-tuning the solving process. In our previous work, we created a high-performance astrometric solver AJAS suited for the Japan Astrometry Satellite Mission for INfrared Exploration (JASMINE). In the present work, we foster AJAS tractability by integrating it with the ontology-driven visual analytics platform SciVi leveraging the principles of multi-purpose ontology-driven API for in-situ data processing. This integration provides users with high-level management tools for AJAS computation jobs and high-level visual data mining tools for AJAS solutions. All these tools can be configured via a graphical user Web interface, extended in Jupyter Notebooks, and executed on the same computing resource as AJAS, which minimises the data transfer. In this paper, we elaborate on the technical details of the above-mentioned tools and demonstrate their capabilities on the real examples of the AJAS solution quality assessment.