The integration of edge machine learning into all-optical fluorescence lifetime-based sensing with nitrogen-vacancy (NV) centers enables efficient, real-time magnetic field measurements. NV center-rich diamonds have demonstrated high precision and non-invasive capabilities for magnetic field detection. By leveraging fluorescence decay characteristics, low-field ambiguities can be resolved and sensing accuracy can be enhanced under varying environmental conditions. This work adapts established NV-based sensing methodologies for a low-cost commercial analyzer platform, implementing lightweight neural networks for real-time inference of magnetic fields. The proposed approach reduces computational overhead and latency, making high-precision magnetic field sensing more accessible for industrial applications. Additionally, challenges related to model optimization and deployment in resource-constrained environments are investigated.