ChatGPT WORK Masterclass

Agentic work modes such as ChatGPT Work reflect exactly the kind of AI Dr. Alfred Ang specialises in: systems that plan and complete multi-step tasks rather than simply answering a prompt. He is a full-stack AI developer and Agentic AI solution architect who designs autonomous and multi-agent systems with frameworks including the OpenAI Agents SDK, the Claude Agent SDK, the Google Gemini Agent SDK, Microsoft AutoGen, CrewAI, n8n, Langflow and Flowise. As Managing Director, CIO, CTO and Chief Instructional Designer of Tertiary Infotech Academy Pte Ltd, he leads agentic AI adoption for both his own organisation and the professionals he trains, helping them move from experimentation to operational AI in customer service, internal workflows, analytics and decision support. His credentials include Microsoft Certified AI Engineer Associate (AI-102) and Data Scientist (DP-100), AWS Certified AI Practitioner, a Certificate in AI Ethics and Governance (Senior Professional) from NTU and a PhD in Electrical Engineering from NUS. His public GitHub portfolio shows how he delegates real work to agents: an AI-powered content management system with a weekly auto-blog scheduler built on the Claude Agent SDK, an automated video production pipeline in which an agent researches and scripts videos from a web-form request, a social media publishing platform that posts to 22 platforms with scheduling and queues, and a set of Claude Code skills that automate tasks such as README generation and LinkedIn posts. He is also a certified PMP, so he thinks of agent tasks in terms of scope, deliverables and review.

That experience shapes a masterclass focused on handing tasks to ChatGPT Work confidently and safely. Learners start by setting up ChatGPT Work and understanding how it differs from standard ChatGPT chat and from Codex, then learn to write clear task briefs that state the goal, inputs, constraints and expected deliverable, and to manage the permissions the agent receives. He guides them through automating research, writing and reporting tasks, and working with files, spreadsheets and connected apps, using realistic office scenarios rather than toy examples. Exercises include compiling a research summary from several sources, turning a folder of documents into a structured report and updating a spreadsheet from connected apps, with attention to where the agent needs clearer instructions or narrower access. The final topic, on reviewing output and building reusable workflows, reflects a principle he applies in his own agent designs: an agent's work should always be checked, and a task that works once should be captured so it can be repeated reliably. His AI ethics and governance training informs practical guidance on data handling, access and accountability. As an ACTA-certified trainer and DACE-certified curriculum developer, he keeps the class code-free and hands-on, so managers, analysts and administrators leave with a set of task briefs and workflows they can use immediately. Learners who want to see his work for themselves are welcome to browse his GitHub portfolio at https://github.com/alfredang and connect with him on LinkedIn at https://www.linkedin.com/in/angchewhoe/