Copilot for Power BI

Making data easy to read is a common thread in 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, from front ends and back ends to databases, cloud deployment and native mobile apps, and many of them centre on reporting. The admin dashboards of the ai-mms course-registration platform, which runs the Singapore, Malaysia and Ghana sites, surface registrations, classes and rosters drawn directly from a MySQL database. Dr. Ang has built project-portfolio dashboards with KPI tiles, status charts and timelines, a warehouse inventory dashboard with derived stock status, and a stock analysis dashboard that combines live market charts with an AI agent interpreting the trends. An inventory forecasting agent running on a local language model calculates projections and flags exceptions automatically, and the HR and accounting platform built for the academy aggregates leave, attendance, payroll and expense data for management. Data modelling skills show up across PostgreSQL, MySQL and SQLite projects, and interactive Python and Streamlit labs teach data mining and machine learning to beginners. A real-time polling app with live charts and a certification practice-exam platform add further experience in capturing and presenting results. All of this work, more than 300 public repositories, is available on GitHub. Each one began with the question of which number a manager actually needs to see.

In Microsoft Copilot for Power BI, Dr. Ang applies that experience of designing reports people actually use. The course starts with getting started with Copilot in Power BI and Microsoft Fabric: what Copilot can do, how it relies on a well-prepared semantic model, and how to write prompts that describe the business question clearly. In the second topic, learners create reports and visuals with Copilot, generating report pages from a prompt, asking for suggested visuals, refining layouts and adding narrative summaries that explain insights in plain language. The third topic covers AI-assisted data modelling and analysis, including naming tables and columns so Copilot understands them, defining relationships, writing and explaining DAX measures with Copilot's help and asking questions of the data to uncover trends and outliers. Dr. Ang stresses validation throughout: checking that a generated measure calculates what the business means, comparing totals against source data and knowing when to build a visual manually. Each concept is demonstrated live on a realistic dataset before learners repeat the exercise themselves, with time to troubleshoot and compare results. Discussions cover licensing and capacity requirements, data security and how to roll out Copilot features to a reporting team sensibly. Learners leave with a completed report, a set of tested prompts and a clear approach to using Copilot for faster, more reliable business intelligence. Learners also practise explaining Copilot-generated insights to colleagues who did not build the report. To see the dashboards and data tools Dr. Ang has built, visit https://github.com/alfredang