3d model reconstruction

I am looking to reconstruct the virtual reality digital twin and 3d model of an space from a video capture. For this, technologies such as Nvidia (instant) NeRF can be used (open to other suggestions if you are an expert in the field). The 3d models is then needed to be transformed into a VR digital twin of the video. I need an experienced freelancer who can work with ML technologies, has access to an Nvidia-GPU equipped device (for running instant NeRF) and is familiar with image processing and 3d model mapping. You need to be able to provide some insights about what works and what needs improvement in this area and help iteratively with improvements of the model to build a high-quality result.