Agentic Software Engineer

Salary: £68,000 - 108,000 per year

Requirements:

• Substantial production software engineering experience, with strong foundations and the judgement to understand architecture, production systems, and risk.
• Regularly delegate substantial engineering work to AI agents, focusing on defining outcomes, providing context, and designing verification rather than directing implementation step by step.
• Systematically increase the scope of work agents can handle independently by improving context, tools, verification, or structure.
• Think in systems about agents environments, access to information, progress through work, and how success is measured.
• Build ways to verify correctness rather than relying on plausible-looking generated output or personally reading everything.
• Work directly with technical and non-technical stakeholders to understand goals and find a pragmatic path to production.
• Learn unfamiliar systems quickly; experience with specific languages or technologies is less important.
• Experience building AI systems may be relevant when the underlying problems transfer to agentic software engineering.

Responsibilities:

• Own outcomes rather than tickets, working with product owners and stakeholders to take problems from idea to production.
• Design workflows in which agents can plan, execute, and verify work against defined outcomes, constraints, and evidence.
• Build and continually improve the agent harness, including instructions, skills, tools, permissions, environments, validation, and feedback loops.
• Design how agents discover and retain relevant context about our architecture, conventions, decisions, product intent, operational state, and previous work.
• Help move work from intent through planning, implementation, verification, deployment, and observation with progressively less synchronous human intervention.
• Enable multiple agent workstreams to run in parallel and converge on tested, shippable outcomes.
• Build tests, evaluations, observability, and feedback systems to establish whether autonomous work is complete and trustworthy.
• Turn successful approaches into reusable capabilities that help the wider team safely take increasingly ambitious ideas into production.

Technologies:

• AI
• AI Agents
• LESS

More:

We are Hult, a global business school that teaches a Computer Science for Business degree. Our engineering team is part of that mission, and how we build software, adopt tools, and approach automation informs what we teach. We offer engineers real ownership, open-ended problems, and a team that trusts them to shape an approach and deliver. We ship quickly and iterate constantly, aiming to shorten the path from idea to production. AI is already helping more of our team ship changes, and we are exploring what AI-native engineering looks like in practice. This is a full-time, on-site role based at our Chelsea office in London, reporting to the Engineering Manager. We ask applicants to answer three short questions about their experience with agents and welcome relevant projects, repositories, experiments, or other examples of their work.

last updated 40 week of 2026