Software Engineer, Machine Learning Infrastructure

Deliveroo
London
Workplace: OnsiteFull timeFunction: Software EngineeringEducation: bachelorsSkills: ["Collaboration","Problem-solving","Operational excellence","Working in ambiguous environments"]

Build production infrastructure for generative AI, focusing on Deliveroo’s open-weights model platform across real-time GPU serving, high-throughput batch inference, and fine-tuning. Work on GPU autoscaling/utilization, inference engines, backend services, and observability to reduce cost and latency while meeting production SLOs. Partner with ML, product, data science, and platform teams across Deliveroo, DoorDash, and Wolt to turn GenAI prototypes into durable platform primitives.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Deliveroo
Deliveroo
1 month ago

Software Engineer, Machine Learning Infrastructure

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 17 hours agoStatus: Live

Job Summary

Build production infrastructure for generative AI, focusing on Deliveroo’s open-weights model platform across real-time GPU serving, high-throughput batch inference, and fine-tuning. Work on GPU autoscaling/utilization, inference engines, backend services, and observability to reduce cost and latency while meeting production SLOs. Partner with ML, product, data science, and platform teams across Deliveroo, DoorDash, and Wolt to turn GenAI prototypes into durable platform primitives.
Location: London
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering
Seniority: Mid level

Key Responsibilities

  • •Build infrastructure to move GenAI ideas from prototype to production and increase AI impact velocity.
  • •Develop and improve the open-weights serving stack (real-time GPU endpoints, batch inference, and fine-tuning via SFT/DPO/LoRA) with supporting platform components like gateways and guardrails.
  • •Design scalable, high-performance systems for model serving, batch inference, GPU autoscaling, and fine-tuning for customer and internal automation use cases.
  • •Improve GPU inference cost/latency by optimizing throughput and reducing end-to-end processing time while ensuring reliability, fallback, and cost controls.
  • •Partner with ML, product, data science, and platform teams across Deliveroo, DoorDash, and Wolt to turn emerging GenAI capabilities into reusable platform primitives.

Key Requirements

  • •BSc, MSc, or PhD in Computer Science (or equivalent).
  • •3+ years of industry experience in software engineering.
  • •Strong backend fundamentals, especially Python and distributed systems.
  • •Experience building production services, APIs, data pipelines, or ML infrastructure at scale.
  • •Hands-on experience with LLM inference and/or fine-tuning of open-weight models in production (serving and/or SFT/DPO/LoRA).
Education:Bachelor's in Computer Science
Skills:CollaborationProblem-solvingOperational excellenceWorking in ambiguous environments
Tech Stack:PythonLLMVLMGPU servingBatch inferenceFine-tuningSFTDPOLoRADistributed systemsLLM GatewayAgent GatewayObservabilityMonitoringSLOsKubernetesAWSGCPVLLMSGLang

Company Brief

Deliveroo
Deliveroo is a multinational online food and grocery delivery marketplace connecting consumers, restaurants, retailers and riders via its app and platform, operating across multiple countries with fast local delivery and logistics services.
Industry: Online Marketplaces
Company Size: Enterprise (1,001+ employees)
Revenue: USD 1B+
Growth: Public Company
Valuation: Unicorn (USD 1B+)
Funding: IPO / Publicly Listed
Headquarters: London, United Kingdom
Founded: 2013
Glassdoor
Glassdoor: 3.4
WebsiteLinkedInGlassdoor