(M/F) Postdoctoral Position: Dielectric Metasurfaces for Single-Photon Sources

id="offer-description" class="ecl-u-type-heading-2 ecl-u-mt-none ecl-u-mb-l">Offer Description • Deep-learning inverse design for multi-scale optimization (meta-atom + metasurface). • Experimental validation using state-of-the-art nano-positioning (LPCNO collaboration). • The CEMES–LAAS synergy: quantum nano-optics + numerical modeling & cleanroom fabrication. The postdoctoral fellow will lead the integration of numerical and experimental workflows, focusing on:. 1. Advanced numerical framework. Extend the autodiff capable torchGDM toolkit (developed at LAAS) to model quantum emitters coupled to periodic metasurfaces supporting BIC modes (array-scanning technique, Fig. 1). Develop physics-informed neural networks to accelerate simulations and inverse design. 2. Deep-learning inverse design. Train multi-objective neural networks to optimize both meta-atom geometries and collective metasurface responses (e.g. simultaneous near-field enhancement and directional emission). Use transfer learning to fine-tune the models on experimental data. 3. Experimental validation & iterative refinement. Characterize the optimized metasurfaces on CEMES optical benches (4K cryostat, time-resolved photoluminescence mapping, HBT single-photon statistics), with precisely positioned quantum emitters (LPCNO, Q-META ANR project). Run iterative cycles: design → fabrication (LAAS cleanroom) → characterization → model refinement. 4. Mentorship & collaboration. Co-supervise PhD students (Q-META, SPINOSA a