Microsoft 365 Copilot for Excel

Spreadsheets, dashboards and the logic behind them appear throughout the work of Dr. Alfred Ang, a full-stack AI developer, Agentic AI solution architect and corporate trainer. As Chief Technology Officer of Tertiary Infotech Academy Pte Ltd, the Singapore training organisation behind Tertiary Courses, Dr. Ang builds solutions end to end, covering front ends, back ends, databases, cloud deployment and native mobile apps. Much of that work involves turning raw tables into decisions. Dr. Ang built IT project management Kanban boards with large KPI tiles, colour-coded statuses and project status charts, a portfolio dashboard with project timelines, and a warehouse inventory dashboard with a filterable, sortable stock table and derived stock status. An inventory forecasting agent running on a local language model reads stock levels, flags items under reorder thresholds and computes forecasts automatically. The HR and accounting platform that Dr. Ang developed for the academy calculates leave balances, attendance hours, payroll and expense claims, the kind of rules-based calculation that many teams still maintain in Excel. Interactive Python and Streamlit labs teach beginners data mining and machine learning, and a stock analysis dashboard combines live charts with an AI agent that interprets price movements. Dr. Ang has also written Claude Code skills that package instructions and rules for repeatable AI tasks, an idea that maps closely onto Copilot's skills and rules. The full collection, more than 300 public repositories, is open on GitHub for anyone to explore.

Microsoft 365 Copilot for Excel follows a progression of eight labs across three topics, and Dr. Ang uses real-world data habits to make each one count. In Excel Fundamentals with Copilot, learners prepare and clean data, write effective prompts, generate and explain formulas, analyse results and create charts, learning to check every formula Copilot proposes rather than accepting it blindly. The second topic introduces skills and rules: reusable instructions that tell Copilot how to format outputs, apply business logic consistently and follow the team's conventions, so the same analysis can be repeated next month without rewriting prompts. Dr. Ang draws on experience building reusable AI skills to show how clear, well-scoped instructions produce reliable results. The third topic brings everything together in interactive dashboards built with Copilot, skills and rules, where learners combine pivot tables, charts, slicers and summary insights into a dashboard that a manager can understand at a glance. Throughout the labs, Dr. Ang demonstrates a task, explains the reasoning behind the prompt and then lets learners practise on their own copy, with time to troubleshoot unexpected results together. Data accuracy, privacy and knowing when to fall back on a manual formula are discussed openly. Learners leave with completed lab workbooks, a dashboard they built themselves and a set of prompts, skills and rules they can reuse at work. Shortcuts for reviewing and correcting Copilot's output are shared along the way. Dr. Ang's dashboards and data tools can be explored at https://github.com/alfredang