Domain:
Digitalisation
Content Type:
Classroom
Audience:
Manager; Senior Officer
Course Code:
CRPME10
Remark:
Learners would need to bring their own GSIB laptops with internet access for this programme or a GoMax iPad machine. (note: SE-cards machines are not suitable)
Learners who do not have GSIB laptops with internet access unfortunately are unable to sign up for this programme at the moment.
Who Should Attend:
You are a Singapore Public Officer who may be an individual contributor or a data business owner. You are keen to learn the power of prompts to get the most appropriate AI responses from generative language models.
Programme Fees and Grants
PSD will subsidise $364.00 of the total programme fee before GST. The grant is valid from 01 Apr 2025 to 31 Mar 2026. Please note that you need to complete the programme in order to qualify for the grant.
Programme Overview
The world of AI has changed the way we work- with increasing interest, use of chatbots and Generative Pre-Trained Transformers (GPTs). How can we leverage these models to help us do our work more effectively? The answer is in giving these models effective prompts
Join us in this programme that was specially designed to provide you with a practical guide to prompt engineering, focusing on the key principles and techniques for designing and implementing effective prompts. You will also have the chance to design prompts for different specific tasks using LaunchPad (a GovTech platform to enable officers to experience AI safely and securely).
Learning Outcomes
- Gain a better understanding of the architecture and key concepts of Large Language Models (LLMs), including generative AI and tools like ChatGPT.
- Recognise the limitations, pitfalls, and ethical considerations of using LLMs, and discern between good use and misuse.
- Grasp the concept of prompt engineering and its critical role in guiding LLM responses.
- Appreciate the principles of good prompt design, including clarity, context setting, and managing output length and verbosity.
- Apply key LLM parameters and techniques for specific use cases (generate, classify, search, cluster, extract, rewrite, summarise).
- Explore advanced prompt engineering techniques, including iterative prompt development, eliciting specific or creative responses, reducing biases and improving reliability.
- Explore real-world applications of LLMs, learn from case studies, and understand best practices for integration.
- Comprehend the future evolution of generative AI, including its potential, key advancements such as Dall.E and Midjourney, and the role of multimodal models.
Last updated: 06 Mar 2025
1. Generative AI and basics of Large Language Models (LLMs)
- Understanding LLMs: Architecture and key concepts
- What are Foundation Models / Large-Language Models
- Temperature Settings
- Zero shot, One shot, Few shots
- Tokens
- Introduction to LaunchPad Chat features and capabilities
- Limitations, pitfalls, and ethical considerations
- Good use and misuse of LLMs
2. Prompt Engineering concepts
- What is prompt engineering?
- Role of prompts in LLMs
- Principles of good prompt design using the CO-STAR approach (from GovTech’s)
3. LLM parameters and techniques
- Temperature setting: Balancing creativity and consistency
- Hallucination: Causes, consequences, and mitigation
- Techniques for specific use cases
- Advanced prompt techniques
- Iterative prompt development
- Eliciting specific or creative responses
- Reducing biases and improving reliability
- Creative prompting
- Chain of thought
- Role-play
4. Using AI Agents To Do Your Work Faster
5. Future and potential of generative AI
6. What’s Next?
8.00 hours
CSC
Session dates:
- 09 Jun (09:00 - 17:00)
$991.90 per participant
(including 9% GST of $81.90)
Apply Now
Add to Watchlist
Watching
(including 9% GST of $81.90)
Registration information:
- Registration closes on 02 Jun 2025
- First-come-first-serve basis.
- Programme fee may vary on different financial year.
Need help? We are glad to help you
Find out how to contact us here.