Frontier AI Engineer: Architect Multi-Agent Systems

Cognizant; City Of London; Hybrid; GBP 90,000 - 130,000; Full time

Job summary

Cognizant in London is seeking a Frontier Engineer to help build AI that moves the world forward. You'll join a lean pod and partner with clients to design, deploy and improve AI-native systems, with freedom to experiment and turn bold ideas into impact.

In this role you will build and orchestrate multi-agent AI systems, design prompts, create RAG architectures, deploy and monitor LLMs, and build AI-embedded applications using modern engineering practices.

Qualifications

• 3–10 years of technical delivery experience with strong AI-tool fluency.
• Hands-on agentic coding and multi-agent orchestration using frameworks such as LangChain, AutoGen or LangGraph.
• Experience designing RAG pipelines and managing vector databases to ground and improve agent accuracy.

Responsibilities

• Build and orchestrate multi-agent AI systems, integrating agent outputs into production environments that solve real business problems.
• Design intent-driven prompts and workflows for AI coding agents, applying human judgment to validate quality, safety and accuracy.
• Create scalable RAG architectures, vector stores and knowledge systems that enable trustworthy and context-aware AI experiences.
• Deploy, monitor and optimize LLMs and AI agents in production, implementing responsible AI guardrails and observability practices.
• Build AI-embedded applications and workflows using enterprise AI platforms, automation tools and modern engineering practices.

Skills

AI tool fluency

LangChain

AutoGen

LangGraph

Prompt engineering

Tools

LangChain

AutoGen

LangGraph

Job description

Cognizant in London is seeking a Frontier Engineer to help build AI that moves the world forward. You'll join a lean pod and partner with clients to design, deploy and improve AI-native systems, with freedom to experiment and turn bold ideas into impact.

In this role you will build and orchestrate multi-agent AI systems, design prompts, create RAG architectures, deploy and monitor LLMs, and build AI-embedded applications using modern engineering practices.