NUS
 
ISS
 
NUS ISS

Deploying Safe and Secure AI Agents

Enabling safe, reliable, and enterprise-ready AI adoption

Overview

Reference No TGS-2026063782
Part of -
Duration 2 days
Course Time
Enquiry Please contact ask-iss@nus.edu.sg for more details.

A secure AI agent is more than a well-crafted prompt. Once an agent can access enterprise data, invoke tools, or take actions on behalf of users, security must be built into its design and not added as an afterthought.

This intensive two-day, hands-on software engineering course focuses on designing, building, testing, and deploying AI agents that behave safely and securely in real-world environments. Rather than treating security as an operational concern, you'll learn how to engineer it directly into your applications through practical design patterns, testing strategies, and layered controls.

Approximately half the course is dedicated to project work. Working in small teams, you'll build a realistic AI agent, explore how it can fail under realistic attack scenarios, implement effective safeguards, and verify that your solutions work.

By the end of the course, you'll leave with practical techniques, reusable code patterns, and greater confidence in developing AI agents that are trustworthy and ready for enterprise deployment.

No cybersecurity background is required. Participants should be comfortable developing applications in Python and have a basic understanding of Large Language Models (LLMs).

This course is part of the Software Systems series offered by NUS-ISS.

Register

Registration Instructions

Self-sponsored Participants

  • Register for the course by clicking on the "Register Now" button above
  • You may refer to the User Guide for Learner

Company-sponsored Participants

  • You will have to be registered for the course by someone from your company who has an account on the LifeLong Learning Portal (L³AP)
  • The person in-charge may register you for the course by:
    • Generating a corporate registration link for you to register for the course
      • After the link is generated, you must:
        1. Log in to L³AP by clicking on the "Register Now" button
        2. Click on the corporate registration link after logging in
      • If you do not follow the above instructions, you will be registered for the course as self-sponsored
    • Registering you for the course backend
      • You will still be required to log in to L³AP and complete your registration by clicking on the "Register Now" button above
  • You may refer your HR/L&D POC to the User Guide for Company

Upcoming Classes

Class 1 14 Sep 2026 to 15 Sep 2026 (Full Time)

Duration: 2 days

Time:
09:00am to 05:00pm

Class 2 01 Oct 2026 to 02 Oct 2026 (Full Time)

Duration: 2 days

Time:
09:00am to 05:00pm


Key Takeaways

By the end of this two-day course, you will be able to:

  • Differentiate AI safety failures from security failures and determine the most appropriate mitigation approach for each.
  • Analyse common attack vectors targeting AI agents and evaluate whether your defences effectively withstand real-world threats.
  • Design layered security controls that continue to protect AI agents even when models behave unexpectedly or are manipulated.
  • Evaluate AI agents at both the model and system levels, turning failures into repeatable automated tests that improve reliability over time.
  • Apply and assess managed guardrails and security controls, understanding both their capabilities and their limitations.
  • Apply practical engineering techniques that improve the safety and trustworthiness of AI agents before deployment.


Who Should Attend

This course is designed for professionals who build, design, or are responsible for delivering AI agent applications.

It is particularly suitable for:

  • Software, backend, and full-stack developers building AI-powered applications
  • AI application developers moving from prototype to production
  • Technical leads and engineering managers responsible for production-ready AI systems
  • Solution architects with hands-on implementation responsibilities
  • Technology professionals seeking practical approaches to building secure AI agents


Pre-requisites

Participants should have:

  • Experience developing applications using Python
  • A basic understanding of Large Language Models (LLMs)
  • An interest in building AI-enabled applications


What to Bring

No printed copies of course materials are issued.
Participants must bring their laptops (participants will not be able to complete the workshop with their tablets) with power charger to access and download course materials.

If you are bringing a laptop, please see below for the tech specs:

 

Minimum

Recommended

Computer and processor

1.6 GHz or faster, 2-core Intel Core i3 or equivalent

1.8 GHz, 2-core Intel Core i3 or equivalent

Memory

4 GB RAM

8 GB RAM

Hard Disk

256 GB disk size

 

Display

1280 x 768 screen resolution (32-bit requires hardware acceleration for 4K and higher)

 

Graphics

Graphics hardware acceleration requires DirectX 9 or later, with WDDM 2.0 or higher for Windows 10 (or WDDM 1.3 or higher for Windows 10 Fall Creators Update).

DirectX 10 graphics card for graphics hardware acceleration

Others

An internet connection – broadband wired or wireless

Speakers and a microphone – built-in or USB plug-in or wireless Bluetooth

A webcam or HD webcam - built-in or USB plug-in

 



What Will Be Covered

  • Designing Secure AI Agents
    • Understand where AI agent risks originate and why prompt engineering alone is not sufficient to secure enterprise AI applications.
    • Learn how to define operational boundaries through application code, control tool access and permissions, and implement practical security controls that reduce the risk of unintended agent behaviour.
  • Guarding, Testing and Verifying AI Agents
    • Learn how to protect AI agents against real-world threats such as prompt injection and untrusted content.
    • Explore layered AI guardrails, human oversight for high-risk actions, and practical approaches to testing and evaluating agent behaviour to improve security, before deployment.
  • Team Project
    • Each afternoon, participants will work in teams to apply the day's concepts to a realistic AI agent.
    • Participants will identify vulnerabilities, implement practical controls, evaluate their design, and verify their improvements through testing, using code built and validated by yourself.


Fees & Subsidies

Fees for 2026

 Full FeeSingaporeans & PRs
(Self-Sponsored)
  Full course feeS$1,800.00S$1,800.00
  ISS Subsidy -S$180.00
  Nett Course FeeS$1,800.00S$1,620.00
  9% GST on Nett Course FeeS$162.00S$145.80

   Total Nett Course Fee Payable,       Including GST

S$1,962.00S$1,765.80

 

Note:

  1. All fees and subsidies are valid from January 2024, unless otherwise advised.
  2. From 1st January 2024, the GST will be increased to 9%.
  3. For corporate run, please contact us directly for further details.


loading

Certificate

The ISS Certificate of Completion will be issued to participants who have attended at least 75% of the course and pass the required assessments.


Preparing for Your Course

NUS-ISS Course Registration Terms and Conditions

Find out more.

NUS-ISS and Learner’s Commitment and Responsibilities

Find out more.

WIFI Access

WIFI access will be made available to participants.

Venue

NUS-ISS
25 Heng Mui Keng Terrace
Singapore 119615

Click HERE for directions to NUS-ISS

In the event of a change of venue, participants are advised to refer to the acceptance email sent one week prior to the commencement date.

Course Confirmation

All classes are subject to confirmation and NUS-ISS will send an acceptance email to participants one week prior to the commencement date. Confirmed registrants are to attend and complete all lectures, class exercises, workshops and assessments (where applicable). Additionally, all responses to feedbacks and surveys conducted by NUS-ISS and its partners must be submitted. All training and assessments will be delivered as described in the course webpage.

General Enquiry

Please feel free to write to ask-iss@nus.edu.sg if you have any enquiry or feedback.



Course Resources

Develop your Career in the Following
Training Roadmap(s)

Please click on the discipline(s) to view the training roadmap of related courses to assess your training needs and goals.

Software Systems

Architecting the backbones of smart cities

Read More

You Might be Interested in...

A+
A-
Scrolltop
More than one Google Analytics scripts are registered. Please verify your pages and templates.