Leadership, Organisational Culture and Human Factors in FinTech
The organisation we choose to build
The module in plain words
You run Harbour Payments for twelve weeks
Harbour is a growing payments company with a good product, an overworked service team, and a small AI agent pilot that already helps handle merchant disputes. The decision on the table is whether to expand that pilot.
Across the semester your group works out what the culture is really like, tries different ways of leading, tests what the work feels like for staff and merchants, looks into a near-miss, negotiates a change and defends what the company should become.
You lead Harbour by building. Your group uses AI coding agents on the terminal to build three working tools, and you defend the last one live. There are no essays.
The question underneath every week: what kind of organisation are we building through the decisions we make today?
Harbour is fictional. Every internal figure, memo and testimony is invented for teaching. The outside cases you read are real, and they are labelled so you can tell the two apart.
There is no required answer about AI. Expanding the pilot, narrowing it, pausing it, or improving the human process instead are all legitimate conclusions. What earns marks is what you build and the judgement inside it: how theory shaped the design, the choices you made while directing the AI, what happened when people used it, and how well you defend it.
Each week opens with This week at the frontier: fifteen minutes, led by the lecturer, on something recent in AI and financial services. What happened, how do we know, whose account is missing, and what should a leader do? It is not assessed and needs no preparation.
Twelve teaching weeks
The assessment: three builds
You are Harbour Payments' leadership team, and you lead by building. Your group uses a terminal AI agent to build three working artefacts: a tool that reads Harbour's culture, a console that keeps a human in control of Harbour's AI agent, and a final system that changes how Harbour works, which you defend live in front of the board, a worker and a merchant.
We assume the AI wrote most of the code. What earns marks is what you build and the judgement inside it: how theory shaped the design, the choices you made while directing the AI, what happened when real people used it, and how well you defend it. The three builds are worth 100% of the module, weighted 10, 20 and 70. There are no essays and no exam.
How the semester runs
Timings are teaching weeks. Exact submission dates, extensions and the return date for final marks are published on Moodle with each brief. Feedback on each CA comes back before the next one is due.
The rubrics: 100 marks each, no hidden criteria
Each CA is marked out of 100 and then weighted: CA1 counts 10%, CA2 20% and CA3 70% of the module. The gate line only confirms the build exists and runs; the rest is judgement.
Groups of four form in week 1, with timetables and access needs in mind, and agree a short charter: meeting times, roles, response times and what happens if things go wrong. The four roles on Harbour's leadership team (chair, operations and service designer, people and customer advocate, evidence and challenge lead) rotate so that everyone holds each one during weeks 1 to 8. Your commits and the contribution table show who built what; unresolved concerns go through NCI's group-assessment procedure.
You build with a terminal AI agent, and the ai-log folder shows how. You stay responsible for everything the AI produces, so check it. A free route always exists. Never paste real customer data, personal data or confidential material into any AI tool.
What goes in each week, as it is built
Every week card carries the same six slots. Briefing is the deck or narrative for the session. Case room holds that week's Harbour evidence and any public case material. Theory gives the bounded readings with page or section locators. Studio is the in-class task, its timings and an accessible alternative. Decision record is the template and how feedback comes back. Revisit is a captioned recording or an equivalent written version.
Week 1's Briefing deck is live. The other slots are not built yet, so each one is marked as coming rather than linked. When a slot fills, it appears on its week card here and on Moodle. Assessment briefs, including the repeat brief, sit beside these slots rather than inside them.