GenAI Infra Engineer — Open-Weights & GPU Inference

Deliveroo; United States; On-site; USD 140,000 - 190,000; Full time

Job summary

Deliveroo is hiring a Software Engineer for Machine Learning Infrastructure focused on Generative AI. You will join a small team building production infrastructure for open-weight models, real-time GPU serving, batch inference, and fine-tuning pipelines.

You’ll collaborate with ML engineers, product teams, and platform squads to deliver scalable, cost-efficient AI capabilities and robust observability, meeting production standards at scale.

Qualifications

• 3+ years of software engineering experience in industry.
• Strong backend and distributed systems fundamentals.
• Production services, APIs, data pipelines, or ML infra experience.
• Observability, reliability, and cost optimization in production.

Responsibilities

• Build production infrastructure for Generative AI at scale.
• Develop open-weights serving, real-time endpoints, and batch inference.
• Work across inference engines, training pipelines, and GPU autoscaling.
• Ensure latency, reliability, and observability meet production standards.

Skills

Python

Distributed systems

LLM inference

Open-weight models

Backend APIs

Data pipelines

Observability

GPU utilization

Education

CS degree

Higher degree (MS/PhD) optional

Tools

Kubernetes

AWS/GCP

GPU infra

TensorRT/ML frameworks

Job description

Deliveroo is hiring a Software Engineer for Machine Learning Infrastructure focused on Generative AI. You will join a small team building production infrastructure for open-weight models, real-time GPU serving, batch inference, and fine-tuning pipelines.

You’ll collaborate with ML engineers, product teams, and platform squads to deliver scalable, cost-efficient AI capabilities and robust observability, meeting production standards at scale.