Generative AI for Manufacturing* (24 Hours)

- This programme is co-organised with Amazon Web Services.
- This programme is in collaboration with the Sectoral AI Centre of Excellence for Manufacturing (AIMfg).
- This module is one of five in the AI in Manufacturing (AIM) programme. Complete all five AIM modules to qualify for an exemption from one Post-Diploma Certificate (PDC) requirement of the Specialist Diploma in Smart Manufacturing at Ngee Ann Polytechnic. Find out more.
Introduction
Artificial Intelligence, particularly Generative AI and Agentic AI, is increasingly transforming how organisations operate across various industries. In modern manufacturing environments, large amounts of data are generated from sensors, production systems, quality inspections, and maintenance records. Extracting actionable insights from this data can be quite challenging.
This 3-day course is designed as a hands-on, application-focused experience for technical and domain professionals who want to move beyond theoretical concepts and into real implementation. It offers a beginner-friendly pathway, taking learners from the foundations of generative AI to building agentic AI prototypes specifically for manufacturing. Participants are expected to have a basic comfort level with cloud-based tools and should possess beginner-level proficiency in Python.
About the Programme
The course equips learners with practical and career-relevant skills to design and implement agentic AI solutions specifically for the manufacturing sector.
- Stage 1: Foundations of Generative AI and Agentic Systems
Participants will explore the foundations of generative AI, including foundation models, embeddings, and prompt engineering, and reinforce them through a prompt-engineering lab and responsible-AI guardrails. They then progress from large language models to agents, and are introduced to tool use, memory, and multi-agent concepts through a hands-on agent-building workshop. - Stage 2: Building Agentic AI Solutions for Manufacturing
Participants will build agentic AI solutions using modern agentic services and frameworks, and apply them to a real industrial use case. Through guided workshops, they implement an end-to-end manufacturing solution (for example, predictive maintenance) for manufacturing on a managed cloud AI platform, such as the AWS platform. - Stage 3: RAG Workflows, Evaluation, and Multi-Agent Diagnostics
Participants will implement a manufacturing-focused Retrieval-Augmented Generation (RAG) workflow in a hosted notebook/lab environment. They will evaluate and tune retrieval methods for generation quality and extend the solution into a simple multi-agent workflow using LangGraph for diagnostics, potential cause analysis, and recommended actions. Learners will have the ability to leverage large language models (LLMs) to apply them to manufacturing use cases without the need for finetuning
The course concludes with a guided mini project in which trainees will work in small groups to build a simple multi-agent agentic AI application for a manufacturing scenario. This project will cover the diagnostic process, potential causes of issues, and suggest recommended checks or actions.
Who Should Attend
This course is ideal for manufacturing and industrial organisations eager to harness the power of generative AI and agentic AI to significantly boost operational efficiency, enhance equipment reliability, assure quality, and improve decision-making capabilities. It caters to a broad audience, including:
- Technical professionals (engineers, developers, and data/automation experts) tasked with bringing innovative prototypes to life.
- Non-technical stakeholders (leaders and process owners) who must assess the feasibility, value, and potential risks of adopting these transformative technologies.

: Full day
: Morning
: Afternoon