Lead AI Architect: Agentic Systems & Observability

Hitachi, Ltd.; Greater London; On-site; GBP 75,000 - 120,000; Full time

Job summary

GlobalLogic is building AI-first enterprise solutions that help global clients transform their digital platforms and products. As an AI Engineer, you'll own the development of the core agentic frameworks, evaluation pipelines, and observability tooling that let it operate with safety, trust, and intelligence at scale.

You won't just be integrating AI into a product; you'll be building the engineering systems that make it trustworthy, with a focus on production readiness, reliability, and

Qualifications

• 5+ years of software engineering experience, including hands-on experience shipping agentic or LLM-backed systems that are used in production.

Responsibilities

• Architect agentic systems: Design and implement multi-agent architectures that the rest of the team builds on top of.
• Practice eval-driven development: Build and own the evaluation harnesses that measure accuracy, hallucination, and performance before anything ships.
• Solve performance at scale: Tackle RAG latency optimization, context-window management, and intelligent information routing under real production load.
• Build observability: Ship tooling that gives the team real-time insight into model behavior and guardrail compliance.

Skills

AI Development Life Cycle

Orchestration

Python engineering

Evaluation

Observability

Distributed systems

Kafka

Cloud-native

Kubernetes

LangGraph

AutoGen

LlamaIndex

CrewAI

Tools

LangSmith

Arize

LLM CI/CD

Kubernetes

Kafka

SQL

Job description

GlobalLogic is building AI-first enterprise solutions that help global clients transform their digital platforms and products. As an AI Engineer, you'll own the development of the core agentic frameworks, evaluation pipelines, and observability tooling that let it operate with safety, trust, and intelligence at scale.

You won't just be integrating AI into a product; you'll be building the engineering systems that make it trustworthy, with a focus on production readiness, reliability, and