Multimodal AI Inference & Serving Engineer

H; Paris (TX); Hybrid; USD 104,000 - 150,000; Full time

Benefits offered by this job

Competitive salary

Career development

Multicultural team

Job summary

H invites a Research Engineer to advance the inference and serving stack for its multimodal, agentic AI models. This hybrid role is based in Paris or London, with in-office presence three days a week and occasional travel.

You will optimize latency, throughput and cost, prototype inference techniques, and collaborate with the Models team on training-time decisions that affect deployment. A strong software background, ML experience, and the ability to communicate across teams are essential.

Qualifications

• Strong software engineering background with production experience.
• Proficient in Python and at least one systems language (Rust, C++, or Go).
• Hands-on experience with deep learning frameworks (PyTorch, JAX) in industry.
• Solid distributed systems fundamentals.
• Experience with modern cloud environments and ML infra (Kubernetes, etc.).
• Knowledge of transformers and multimodal architectures.

Responsibilities

• Build and operate the inference stack for multimodal agentic models.
• Improve latency, throughput, and cost of model serving.
• Research inference techniques for agent workloads.
• Collaborate with Models team on training-time decisions.
• Integrate inference into AI products with cross-functional teams.
• Evaluate platforms and communicate findings to stakeholders.
• Stay current with advancements in inference and accelerator tech.

Skills

Strong software engineering

Python

Rust

C++

Go

Deep learning frameworks

Distributed systems

Cloud infra

ML model serving

Education

Advanced degree (PhD/MS) with research output

Tools

vLLM

SGLang

TensorRT-LLM

CUDA

Job description

H invites a Research Engineer to advance the inference and serving stack for its multimodal, agentic AI models. This hybrid role is based in Paris or London, with in-office presence three days a week and occasional travel.

You will optimize latency, throughput and cost, prototype inference techniques, and collaborate with the Models team on training-time decisions that affect deployment. A strong software background, ML experience, and the ability to communicate across teams are essential.