FinRisk system is an innovative solution designed to improve risk assessments for investment portfolios. Leveraging state-of-the-art Deep Learning techniques, the system offers advanced risk metrics calculations, including Value-at-Risk and Expected Shortfall. Furthermore, FinRisk offers qualitative risk insights through the analysis of market sentiment of financial assets, utilizing Natural Language Processing algorithms.  Additionally, the system offers a digital twin for what-if-analysis, enabling quantification of portfolio risk changes from potential trading positions.