Multimodal ML Engineer - Production-Ready AI

White Circle; Paris; Híbrido; EUR 90.000 - 130.000; Jornada completa

Ventajas ofrecidas por este puesto de trabajo

Relocation package

Hybrid Paris work

Medical insurance (France-based)

Team off-sites

Descripción de la vacante

White Circle is building an AI safety platform and seeks a Multimodal ML Engineer to train and ship vision, audio, video, and speech models for scalable safety policies. You will design experiments, build data pipelines, and optimize MoE architectures for production-ready inference.

You will work on end-to-end model deployment, evaluation benchmarks across visual QA, video understanding, and audio comprehension, with a hybrid Paris-based role and relocation support.

Formación

• 3+ years training large-scale deep learning models in multimodal domains.
• Strong PyTorch skills with hands-on distributed training experience (DeepSpeed, FSDP, or similar).
• Deep experience with multimodal architectures — encoders, projectors, and LLMs fit together (e.g., vision/audio encoders plus LLMs).
• Hands-on with RLHF/alignment for multimodal: GRPO, DPO, reward modeling.
• Experience with video and/or audio sequence modeling and streaming inference.

Responsabilidades

• Train and fine-tune large-scale multimodal models from scratch and from pretrained checkpoints.
• Extend models across modalities: image understanding, video temporal modeling, long-context processing, and 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, including synthetic data generation.
• Train and optimize MoE architectures for efficient multimodal inference.
• Build alignment pipelines: SFT, DPO, GRPO, reward modeling across modalities, not just text.
• Optimize models for production: quantization, distillation, batching, streaming and low-latency serving.
• Deploy models end-to-end: from research checkpoint to production serving.
• Define evaluation metrics and benchmarks that matter for the product: visual QA, spatial reasoning, video comprehension, speech and audio understanding.

Conocimientos

PyTorch

Distributed training

Multimodal architectures

RLHF / alignment

Video/audio sequence modeling

MoE architectures

Data pipelines

Experiment design

Production optimization

Engineering fundamentals

Herramientas

DeepSpeed

FSDP

MoE

Descripción del empleo

White Circle is building an AI safety platform and seeks a Multimodal ML Engineer to train and ship vision, audio, video, and speech models for scalable safety policies. You will design experiments, build data pipelines, and optimize MoE architectures for production-ready inference.

You will work on end-to-end model deployment, evaluation benchmarks across visual QA, video understanding, and audio comprehension, with a hybrid Paris-based role and relocation support.