Description:
This is an architecture role, not a pure engineering role — but you'll need to get hands-on with the technologies you're evaluating. You'll be expected to test things yourself, form your own technical opinions, and contribute to architecture decisions that most people don't get near until much later in their careers.
We will count on you to:
- Contribute to AI platform architecture — working with the rest of the AI CoE to design how AI capabilities fit together across Marsh's technology landscape
- Identify technologies for evaluation — research emerging AI technologies (LLMs, memory systems, agentic frameworks, guardrails) and recommend which should come up for the next proof-of-concept
- Collaborate on proofs-of-concept — pair with a software engineer on focused PoCs, contributing the architecture perspective: how does this technology integrate with what we have, and where we're going?
- Build reference architectures — contribute to high-level integration diagrams, reference architectures, and technology radar documents that guide AI adoption across the business
- Document architecture for humans and AI agents — produce architecture documentation in formats that are useful to both human engineers and AI coding agents
- Present at architecture forums — contribute to presentations at Marsh's Architecture Council, Dev Council, and business unit forums
- Understand the enterprise landscape — learn Marsh's existing enterprise architecture so that your recommendations account for where we are today, not just where we want to be
What you need to have:
- A Batchelor's degree (Master's preferred) in computer science, software engineering, data science, information systems, or a related discipline
- Up to 2 years of professional experience (including internships and placements)
- Genuine interest in system design and architecture — you think about how things fit together, not just how to make them work in isolation
- Strong intellectual curiosity and high standards — you actively track emerging technology, learn quickly, and care about doing excellent work
- Familiarity with cloud platforms (AWS, Azure, or GCP)
- Hands-on comfort with AI coding platforms and copilots (GitHub Copilot, Claude Code, Cursor, or similar) and the modern languages they support
- Strong communication skills, in writing and in person