Useful intelligence.
Practical applications.
AI creates value when it improves a decision or changes a workflow. My work starts with the operating question, then combines the right information, analysis and software to address it.
A clearer view of performance.
Management information is often spread across financial reports, budget files, sales presentations and market research. I use AI-assisted development and analysis to bring these sources into executive dashboards and focused decision tools.
What this looks like in practice
- Performance views connecting budgets, actuals, commercial indicators and market evidence.
- Profitability analysis by brand, product, channel and region, with consistent definitions.
- Source checks that distinguish a missing feed, a stale period and a genuinely new result.
The discipline: a dashboard is only useful when its numbers can be traced and its limits are visible. A working tool does not make an incomplete source complete.
Make the assumptions visible.
I use AI-assisted research and modelling to organise diligence, compare scenarios and prepare investment discussions. The objective is a clearer view of what supports the thesis, what could break it, and what still needs validation.
- Market and company research translated into structured diligence questions.
- Interactive valuation and return scenarios, with explicit operating and financing assumptions.
- Investment committee narratives that connect commercial logic, economics and execution risk.
The discipline: scenario outputs are conditional calculations. They are not forecasts, transaction commitments or substitutes for primary diligence.
Software shaped by the business.
In independent Swiss watchmaking, I have used AI-assisted development to build a commercial and operations workspace connecting customer relationships, product references, orders and production planning.
- Customer and product views with search, filters and linked records.
- Order books and secured-backlog reporting that separate commercial status from delivery planning.
- Permissioned editing and field-level change history, with controls for concurrent changes.
The discipline: commercial reporting is distinct from accounting. Finance integration and reconciled financial reporting require their own acceptance process.
Human accountability
by design.
- Begin with a decision. Define the question, the user and the intended action before choosing a tool.
- Keep the evidence traceable. Preserve sources, periods and definitions. Label assumptions and unresolved gaps.
- Design the controls. Give the right people access, retain change history, and make review part of the workflow.
- Measure actual benefit. A built tool is evidence of capability. Financial or productivity gains require a baseline and observation.
Questions I am often asked.
How does AI fit with executive leadership?
It supports the work of diagnosing performance, evaluating choices and coordinating execution. The judgement and accountability remain with management.
What kinds of tools have I worked on?
Executive dashboards, profitability and scenario analysis, investment research and models, and custom commercial and operations software.
What makes an AI business application credible?
A defined business use, reliable inputs, visible assumptions, appropriate controls and evidence that the output is useful to its intended user.