Event 2
“Cloud Mastery 2026 #1 AI from Scratch”
Event Objectives
- Basic introduction to Strands Agent.
- Guide on using AI effectively and improving LLM output quality.
- Demo of an IoT AI Project and how to use AWS services.
Speaker Details & Presentation Topics
1. Mr. Vinh Banh Cam - Data Engineer
- Topic: Introduction and Application of Strands Agent
- Main Content: Delved into the concepts and architecture of Strands Agent. Presented how Autonomous Agents can be programmed to execute complex task sequences, automate workflows, and optimize decision-making systems in enterprise environments.
2. Mr. Thinh Nguyen Tuan - DevOps Engineer
- Topic: Prompt Engineering & LLM Output Optimization
- Main Content: Shared advanced Prompt Engineering techniques to maximize the power of Large Language Models (LLMs). Guided how to design structured prompts (such as Retrieval-Augmented Generation (RAG), Role Prompting, and Chain-of-Thought) to control formatting, minimize AI hallucination, and improve the quality and accuracy of the output.
3. Mr. Dinh Le Hoang - AI Engineer
- Topic: Integrating IoT AI with Basic AWS Services
- Main Content: Presented how to combine IoT devices with AI edge computing while transmitting and processing data through core AWS services (such as AWS IoT Core, API Gateway, and Lambda) efficiently and cost-effectively, while maintaining high scalability.
Key Highlights
1. Presentation by Mr. Vinh Banh Cam: Introduction and Application of Strands Agent
- Discussed the limitations of LLMs and how to connect LLMs with external tools.
- Explained what an AI Agent is and why we need it.
- Introduced Strands Agents and their workflow.
- Provided a simple demo to understand how to implement Strands Agents.
2. Presentation by Mr. Thinh Nguyen Tuan: Prompt Engineering & LLM Output Optimization
- Explained why Prompt Engineering is important.
- Discussed what makes a good prompt.
- Emphasized that understanding the cost of AI models is crucial.
- Covered advanced techniques (CoT, ToT, RAG).
- Demonstrated Proptimizer, an extension for prompt optimization.
3. Presentation by Mr. Dinh Le Hoang: Integrating IoT AI with Basic AWS Services
- Introduced the core idea and the problem: club members borrowing items from lockers.
- Created an IoT product capable of scanning faces and logging who borrowed the items.
- Utilized Arduino, cameras, LCD displays, and other hardware components.
- Highlighted the critical role of IoT Core in connecting Lambda, S3, and DynamoDB, forming a highly scalable architecture.
- Video Demo.
Event Experience
- Attending this event helped me visually understand how to combine AI and Cloud in practice.
Learning from highly skilled speakers
- Grasped the standard mindset for designing prompts instead of just typing regular text queries to AI.
- Understood how AI Agents operate to automate tasks.
Practical technical experience
- Directly observed the IoT system scanning faces and saving borrowing logs.
- Clearly saw the incredibly fast data flow from the hardware (Arduino) to the Cloud (AWS IoT Core, Lambda).
- Discovered the Proptimizer tool to assist in writing better prompts.
- Visualized how to use Strands Agent to connect AI with external tools.
Networking and exchange
- Had the opportunity to converse with senior peers to learn from their real-world project experiences.
- Listened to practical questions from attendees, which helped expand my product-building mindset.
Key takeaways
- Need to pay close attention to costs when using AI; writing good prompts helps save a significant amount of money.
- Cloud architecture combined with IoT is not overly difficult if using the right AWS services, and it is easily scalable when there are many users.
Event Photos
Overall, the event not only provided technical knowledge but also helped me change my mindset regarding application design, system modernization, and more effective cross-team collaboration.