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

EURAXESS Ireland; Toulouse; Hybrid; EUR 42,000 - 54,000; Full time

Job summary

CNRS Laboratoire d'analyse et d'architecture des systèmes in Toulouse invites a postdoctoral researcher to lead the integration of numerical and experimental workflows for quantum metasurfaces. The role focuses on deep learning–driven inverse design, multi-scale optimization, and rigorous validation with state-of-the-art nano-positioning and photoluminescence techniques.

The candidate will develop neural networks, extend autodiff toolkits, and supervise PhD students across the Q-META projects,

Qualifications

• PhD in physics, nanophotonics or a related field.
• Strong background in at least two of: numerical modeling, deep learning for inverse design, experimental nano-optics.
• Fluent English; French a plus but not required.

Responsibilities

• Lead the integration of numerical and experimental workflows.
• Develop physics-informed neural networks to accelerate simulations and inverse design.
• Train multi-objective neural networks for metasurface optimization and transfer learning on experimental data.
• Mentor PhD students and collaborate across CEMES, LAAS and LPCNO.

Skills

Deep learning

Numerical modeling

Quantum optics

Experimentation

Education

PhD in physics or nanophotonics

Tools

FDTD

FEM

GDM/CDA

PyTorch

TensorFlow

MATLAB

Job description

Organisation/Company CNRS Department Laboratoire d'analyse et d'architecture des systèmes Research Field Engineering Physics Technology Researcher Profile First Stage Researcher (R1) Application Deadline 25 Nov 2026 - 23:59 (UTC) Country France Type of Contract Temporary Job Status Full-time Hours Per Week 35 Offer Starting Date 1 Mar 2027 Is the job funded through the EU Research Framework Programme? Not funded by a EU programme Is the Job related to staff position within a Research Infrastructure? No

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 and AIM ANR projects) on numerical and experimental tasks.

Single-photon sources generate and transmit indistinguishable photons — the building blocks of quantum information — and are essential for quantum communications. Nano-optics paves the way towards compact, integrated devices that trap, store, release and control single-photon emission: its direction, polarization and spectral content.

Individual nanoantennas already control the photodynamics of quantum emitters, but their limited degrees of freedom have led to a performance plateau. Q-META will break this plateau: we will design, fabricate and characterize high-index dielectric metasurfaces whose AI-optimized periodic patterns host high-Q lattice modes — bound states in the continuum — to gain unprecedented control over accurately positioned quantum emitters.

The project builds on the Q-META ANR initiative and leverages:

• 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

Key references from the team

• Cuche et al., Nanoscale 2023 (QE positioning via AFM nanoxerography).
• Wiecha et al., ACS Photonics 2022 (deep learning for aperiodic metasurfaces).
• Chai et al., ACS Photonics 2023 (BIC modes in metasurfaces).

Why join Q-META?

• Cutting-edge research at the intersection of quantum optics, nanophotonics and AI.
• World-class facilities: CALMIP GPU supercomputer, state-of-the-art optics lab, LAAS cleanroom.
• Collaborative environment across CEMES, LAAS and LPCNO.
• Mentorship and leadership opportunities in a growing field.
• Toulouse's vibrant quantum and AI ecosystem (QuET31, C'NANO, GDR Quantum Technologies, ANITI, ...)

The postdoc will be taking place 50/50 between LAAS and CEMES labs at Toulouse (9 + 9 months)

Requirements

• PhD in physics, nanophotonics, quantum optics or a related field.
• Strong background in at least two of:
  • Numerical modeling of nanostructures (FDTD, FEM, GDM/CDA or similar).
  • Deep learning for inverse design or computational nanophotonics.
  • Experimental nano-optics (confocal microscopy, single-photon characterization).
• Programming: Python (PyTorch/TensorFlow, NumPy, SciPy), MATLAB or equivalent.
• Publication record demonstrating high-impact research.
• Language: fluent English (French is a plus, not required).