Multimodal ML Engineer — Vision, Audio & Video AI

White Circle; Paris; Hybride; EUR 90 000 - 140 000; Plein temps

Avantages offerts par ce poste

Hybrid work model

Relocation package for Paris

Best medical insurance in France

Learning and development support

Résumé du poste

White Circle is hiring a Multimodal ML Engineer to train and ship vision, audio, video, and speech models for an AI safety platform handling 100M+ API calls monthly. You will build end-to-end multimodal systems and optimize for production readiness.

The role requires 3+ years of experience with multimodal DL, PyTorch, and distributed training. This position offers hybrid work from Paris or London, with relocation support for Paris and strong medical insurance.

Qualifications

• 3+ years of experience training large-scale multimodal DL models.
• Strong PyTorch skills with distributed training experience (DeepSpeed, FSDP, or similar).
• Experience with multimodal architectures integrating vision, audio, and language models.
• Hands-on RLHF/alignment for multimodal: GRPO, DPO, reward modeling.
• Experience with video and audio sequence modeling and streaming inference.
• Track record of shipping models to production with latency targets.

Responsabilités

• Train and fine-tune large-scale multimodal models (vision-language, audio, speech) from scratch and pretrained checkpoints.
• Extend models across modalities: image understanding, video temporal modeling, long-context processing, streaming audio.
• Design and run experiments: architecture changes, data mixes, training recipes.
• Build and maintain multimodal data pipelines from raw images, video, and audio to training-ready datasets.
• Train and optimize MoE architectures for efficient multimodal inference.
• Build alignment pipelines: SFT, DPO, GRPO, reward modeling across modalities.
• Optimize models for production: quantization, distillation, batching, streaming, low-latency serving.
• Deploy models end-to-end from research checkpoint to production serving.
• Define evaluation metrics and benchmarks relevant to the product.

Connaissances

Multimodal DL

PyTorch

Distributed training

MoE architectures

RLHF alignment

Video/audio sequence modeling

Production deployment

Data curation

Engineering fundamentals

Outils

DeepSpeed

FSDP

LLaVA

Qwen-VL

Whisper

Audio Qwen

Conformer

Description du poste

White Circle is hiring a Multimodal ML Engineer to train and ship vision, audio, video, and speech models for an AI safety platform handling 100M+ API calls monthly. You will build end-to-end multimodal systems and optimize for production readiness.

The role requires 3+ years of experience with multimodal DL, PyTorch, and distributed training. This position offers hybrid work from Paris or London, with relocation support for Paris and strong medical insurance.