The session began with a warm welcome and sincere greetings to all participants by the IEEE Sensors Council, which clearly explained the purpose and objectives of organizing HelloGPT. The club highlighted the importance of gaining awareness of emerging AI technologies, encouraging hands-on learning, and enabling participants to develop the skills required to interact effectively with AI systems.

Zulqarnain Mohammed, Chairperson, opened the session with a brief introduction, outlining the overall structure of the program and emphasizing the key learning outcomes. This introduction helped participants understand the goals of each segment and prepared them for active engagement throughout the session.

The first segment was presented by Samia Rahman, Web Master, who delivered a comprehensive introduction to Artificial Intelligence (AI) and Large Language Models (LLMs) such as ChatGPT, Gemini, and Claude. She explained the concept of AI and how LLMs are trained on large datasets to identify patterns and generate coherent, contextually relevant text. The session clearly distinguished between training, which involves teaching the model using data, and prompting, which focuses on giving instructions to obtain specific outputs. Samia also discussed the role of prompt engineers in real-world applications and demonstrated how natural language is converted into machine-understandable instructions, effectively connecting theoretical concepts with practical use cases.

The second segment, conducted by Zubair Ahmed Khan, Joint Secretary, focused on the fundamentals of prompt engineering. He explained what a prompt is and emphasized how well-structured instructions can greatly improve the accuracy and relevance of AI-generated responses. He introduced the anatomy of an effective prompt, highlighting four essential components: Context, Task, Constraints, and Examples. He stressed the importance of clarity, specificity, and aligning the prompt with the intended tone, role, and audience. To reinforce these concepts, he demonstrated the difference in outcomes between a well-crafted prompt and a poorly written one.

The third segment was delivered by Mohammed Omer, Execom Member, who discussed various prompt types and advanced prompting techniques. He explained how role-based prompts assign specific personas to guide AI behavior, while instruction-based prompts provide clear, task-focused directions. He highlighted the effectiveness of example-based prompts in maintaining tone and style and demonstrated chain-of-thought prompting to encourage step-by-step reasoning. Additionally, he introduced zero-shot and few-shot prompting, illustrating how the inclusion or absence of examples affects accuracy. Through practical demonstrations, he showed how each technique influences AI responses and supports diverse user goals.

The hands-on segment, facilitated by Chairperson Zulqarnain Mohammed, actively engaged participants through a series of interactive, prompt-based activities. The session began with a CTCE rewriting exercise, where participants improved poorly structured prompts by incorporating Context, Task, Constraints, and Examples to achieve clearer and more precise AI outputs. This was followed by a creative prompt challenge in which participants composed a three-line horror story containing specific elements, such as a door and a clock, demonstrating how constraints can guide creativity. The final activity introduced prompt chaining, allowing participants to create multi-step prompts and observe how sequential instructions enhance coherence and output quality. These practical activities provided participants with first-hand experience in designing, refining, and optimizing prompts for real-world applications.

The fifth and final segment was presented by Sara Farzana, Execom Member, who introduced the concept of Reverse Prompt Engineering. She explained that this technique involves analyzing AI-generated outputs to identify the underlying instructions, structure, and reasoning patterns that influenced them. Sara discussed its practical applications in improving prompt quality, increasing model accuracy, and gaining a deeper understanding of how AI interprets natural language. She also demonstrated clear examples of reverse prompting, showing participants how to break down an output, identify missing or implied context, and reconstruct a more effective prompt based on those insights.

The session concluded with an engaging quiz conducted by the Chairperson, which assessed participants’ understanding of the topics covered and provided an energetic and interactive conclusion to the overall learning experience.