Staff AI Engineer

Multiverse
Edinburgh
Workplace: HybridFull timeFunction: Data Science & Machine LearningSkills: ["Communication","Technical leadership","Mentorship","Product thinking"]

Own the architecture and production execution of an internal agentic operating system. Build and ship multi-agent, agent-orchestrated systems that integrate with existing Multiverse infrastructure via APIs and MCP-style interfaces. Design coordination patterns, evaluation/quality infrastructure, and cost engineering (token economics, caching, routing). Drive reliability and standards across prompt management, guardrails, and testing while providing technical mentorship within a small, fast-moving squad.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
Multiverse
Multiverse
5 days ago

Staff AI Engineer

✓ 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

Own the architecture and production execution of an internal agentic operating system. Build and ship multi-agent, agent-orchestrated systems that integrate with existing Multiverse infrastructure via APIs and MCP-style interfaces. Design coordination patterns, evaluation/quality infrastructure, and cost engineering (token economics, caching, routing). Drive reliability and standards across prompt management, guardrails, and testing while providing technical mentorship within a small, fast-moving squad.
Location: Edinburgh
Workplace: Hybrid
Employment Type: Full time
Job Function: Data Science & Machine Learning
Seniority: Sr. Manager level

Key Responsibilities

  • •Own the technical architecture of an internal agentic operating system, including orchestration, context strategy, tool integrations, evaluation, and production operations.
  • •Ship production AI agent systems by writing code, reviewing code, and delivering significant agent systems end-to-end.
  • •Design multi-agent coordination patterns (task decomposition, handoff protocols, shared state management, orchestration logic).
  • •Build evaluation and quality infrastructure (automated eval pipelines, human-in-the-loop review, regression testing, quality metrics).
  • •Drive cost engineering and integration layers (token economics, caching, model routing, prompt optimisation; APIs/shared data contracts that make existing systems agent-accessible).

Pay and Benefits

Perks:Paid LeaveHealth InsuranceGym Membership

Key Requirements

  • •Have shipped multi-agent systems or complex AI products to real users, with production experience in agent engineering.
  • •Expertise in context management (retrieval, chunking, memory/summarisation) and the cost/quality trade-offs of production designs.
  • •Experience with model selection and routing logic for capability, latency, cost, and reliability.
  • •Ability to build tool use/agent augmentation layers (tool descriptions, failure handling, MCPs or equivalent interfaces).
  • •Build eval and quality infrastructure (automated pipelines, human-in-the-loop review, regression testing) before shipping agent systems.
Experience:EdTech
Skills:CommunicationTechnical leadershipMentorshipProduct thinking
Tech Stack:Claude CodeMCPsAPIs

Company Brief

Multiverse
Multiverse provides apprenticeship and upskilling programs combining human coaching with AI to train employees in data, tech and AI skills, partnering with large organisations to deliver on-the-job learning and career mobility.
Industry: Corporate Training
Company Size: Enterprise (1,001+ employees)
Revenue: USD 50M to 100M
Growth: Scaleup
Valuation: Unicorn (USD 1B+)
Funding: Series D
Headquarters: London, United Kingdom
Founded: 2016
Glassdoor
Glassdoor: 2.9
WebsiteLinkedInGlassdoor