The presentation by Intelligent Light delves into leveraging digital engineering to enhance computational analyses. They advocate for a stochastic basis in CFD, suggesting the replacement of a single solution with an ensemble of solutions. This approach considers various factors like boundary conditions, discretisation error, solver parameters, and the CFD code itself. The goal is to drive towards quantities of interest with confidence intervals, such as estimating the maximum temperature within a specific confidence range. Key goals outlined include achieving more certainty and automation in CFD UQ. These encompass handling many sets of results, ensuring interoperability of data models, accrediting models/results as a single source of truth, increasing the fidelity of physics models, and utilising multi-fidelity approaches from high-fidelity CFD to surrogate models, supported by good test/experimental reference data. The presentation addresses several issues and ideas, such as how HPC can enable more simulation ensembles, the automation of workflows, data management, building and training surrogate models, operating in a multi-fidelity environment under uncertainty, and tracking vast amounts of metadata and result artifacts for reuse and customer audit. Intelligent Light proposes a holistic approach to these challenges, suggesting the use of an open toolset to orchestrate workflows, collect metadata, and record provenance. Integrating data productivity directly with solver codes and making sensitivity analysis and UQ a standard part of workflows are also recommended. The conclusion emphasizes that unique configurations and 'margin squeeze' in engineering methods can be addressed by HPC and new AI surrogate techniques. Managing metadata and provenance presents benefits for businesses and customers. However, the biggest hurdle identified is the resistance to change within organizations. The presentation concludes with thanks and recognition of the transformative tools available for engineering.