AI Research Intern: Architecture Diffusion with Equity

Meetdavis; Paris; Sur place; EUR 13 000 - 20 000; Temps partiel

Résumé du poste

Davis is seeking an AI Research Engineer intern to drive groundbreaking work on a foundation model for architectural design in Paris. You will work with a small team of engineers and architects to push the state-of-the-art in discrete diffusion for floorplans, while building scalable data pipelines and evaluating novel perturbations.

You will contribute to model architecture exploration, training on multi-GPU nodes, and data curation, with emphasis on reproducibility and rigorous ablations.

Qualifications

• Currently in a top MSc/PhD in ML, CS, applied maths, or a closely related field.
• Experience with diffusion models or generative modeling.
• Strong Python and PyTorch programming skills.
• Experience with distributed training / PyTorch Lightning.
• Excellent collaboration and communication.

Responsabilités

• Model Architecture & Design Space exploration: read literature, form hypothesis, design ablation, run it, and interpret results.
• Model improvement: training on multi-GPU nodes, analyze loss curves, and iterate on architecture and conditioning.
• Dataset creation: build a pipeline that parses plans into a clean, structured representation.

Connaissances

Applied Research

Diffusion Models

Python & PyTorch

Research Literacy

Collaboration

Formation

MSc/PhD candidate in ML/CS/applied maths

Outils

PyTorch

PyTorch Lightning

Multi-GPU Training

Description du poste

Davis is seeking an AI Research Engineer intern to drive groundbreaking work on a foundation model for architectural design in Paris. You will work with a small team of engineers and architects to push the state-of-the-art in discrete diffusion for floorplans, while building scalable data pipelines and evaluating novel perturbations.

You will contribute to model architecture exploration, training on multi-GPU nodes, and data curation, with emphasis on reproducibility and rigorous ablations.