AI Architect: Secure, Scalable Agent Systems

Mott MacDonald

Manchester

Hybrid

GBP 80,000 - 120,000

Full time

1 hour ago
Be an early applicant

A complete application in a minute

Tailored resume and cover letter, ready to send.

Get past ATS filters

Benefits offered by this job

Private medical insurance
Pension plan
Employee ownership
Flexible benefits
Wellbeing program

Job summary

Mott MacDonald, a global engineering and advisory firm, is expanding its Central Digital AI team in the UK. This hands-on role shapes secure, governed AI and agent systems, owning reference architectures and patterns that scale across projects.

You’ll collaborate with IT, cyber security and enterprise architecture to test prototypes, define cost-aware solutions and produce architecture documentation for governance and risk assessments.

Qualifications

  • Experience architecting production cloud solutions, preferably on Microsoft Azure.
  • Hands-on knowledge of LLM and agent application patterns: RAG, tool calling, multi-agent orchestration, context engineering and evaluation.
  • Understands how LLM and agent systems fail, and how to design controls, fallbacks and approval points around them.
  • Good understanding of integration patterns, APIs and data platforms.
  • Able to build prototypes to validate a design. Comfortable writing, debugging and maintaining code by hand, and effective at using AI coding agents to accelerate delivery while critically reviewing what they produce.
  • Communicates architecture clearly, in writing and in person, to technical, security, governance and non-technical audiences.

Responsibilities

  • Own the reference architectures for AI systems, including agent harnesses, multi-agent orchestration, tool and MCP integration, retrieval (RAG), memory and context management, model hosting and data access.
  • Define model-agnostic patterns so foundation models can be swapped or upgraded without re-engineering solutions, with routing between models based on cost, latency and capability.
  • Separate what agents can attempt from what they are allowed to do: models plan and request actions, while a trusted control layer handles identity, authorisation, policy, approvals, tool execution and audit, in line with data classification.
  • Structure systems so that people and agents can work in parallel, with clear contracts and interfaces between components and enforced dependency rules, so many changes can proceed at once and components can be replaced without reworking the rest.
  • Review solution designs from the team and projects, and keep a traceable record of architecture decisions.
  • Work with IT, cyber security, enterprise architecture and platform teams to keep AI solutions aligned with enterprise standards and change management.
  • Assess build-versus-buy and model or platform choices on capability, cost, supportability and lock-in.
  • Design observability, tracing, logging and audit into AI and agent systems, meeting the traceability requirements agreed with the responsible AI and governance function.
  • Produce the technical documentation needed to support AI risk assessments and approvals.
  • Coach engineers and data scientists on architecture and design, and support bids with solution architectures and effort estimates.

Skills

Cloud architecture
Azure
LLM patterns
Agent orchestration
Security compliance

Tools

Azure Databricks
APIs integration

Job description

Mott MacDonald, a global engineering and advisory firm, is expanding its Central Digital AI team in the UK. This hands-on role shapes secure, governed AI and agent systems, owning reference architectures and patterns that scale across projects.

You’ll collaborate with IT, cyber security and enterprise architecture to test prototypes, define cost-aware solutions and produce architecture documentation for governance and risk assessments.

Get your free, confidential resume review.

or drag and drop your file here.

Similar jobs

Similar jobs worth comparing

AI Architect: Secure, Scalable Agent Systems
AI Architect: Secure, Scalable Agent Systems

Mott MacDonald • Greater London

Hybrid
GBP 90,000 - 130,000
Private medical insurance
Employee ownership
Flexible benefits
+2
AI Architect — Cloud, Governance & Azure
AI Architect — Cloud, Governance & Azure

Mott MacDonald • City of Edinburgh

Hybrid
GBP 90,000 - 140,000
Private medical insurance
Pension scheme with employer match
Employee ownership
+2
Azure AI Architect - Secure, Scalable & Governed
Azure AI Architect - Secure, Scalable & Governed

Mott MacDonald • Liverpool

Hybrid
GBP 90,000 - 120,000
Private medical insurance
Life assurance up to 4x basic salary
Income protection
+5
AI Engineer (UK-wide, Hybrid) - Multimodal & Production ML
AI Engineer (UK-wide, Hybrid) - Multimodal & Production ML

Mott MacDonald • Bristol

Hybrid
GBP 70,000 - 110,000
Private medical insurance
Hybrid working
Excellent pension scheme
+1
AI Delivery Lead: Automations & Agents for Major Projects
AI Delivery Lead: Automations & Agents for Major Projects

Mott MacDonald • Manchester

Hybrid
GBP 90,000 - 130,000
Private medical insurance
Hybrid working policy
Employee ownership benefits
+1
AI Architect
AI Architect

AutoThink Group • Rochdale

Hybrid
GBP 120,000 - 180,000
AI Systems Architect - GenAI/LLM, Multi-Agent Orchestration
AI Systems Architect - GenAI/LLM, Multi-Agent Orchestration

HM Revenue & Customs • United Kingdom

On-site
GBP 90,000 - 140,000
AI Delivery Lead: End-to-End & Scalable Solutions
AI Delivery Lead: End-to-End & Scalable Solutions

Mott MacDonald • Greater London

Hybrid
GBP 90,000 - 120,000
Private medical insurance
Pension contributions match (4.5%–7%)
Flexible benefits programme
AI Platform & Infrastructure Lead — Secure Cloud & MLOps
AI Platform & Infrastructure Lead — Secure Cloud & MLOps

Mott MacDonald • City Of London

Hybrid
GBP 70,000 - 100,000
Private medical insurance
Pension scheme and life assurance
Flexible benefits programme
+1
AI Architect
AI Architect

Cognitive Group | Part of the Focus Cloud Group • Greater London

Remote
GBP 120,000 - 160,000