Punsiri Boonyakiat
Senior Data Engineer
Bangkok, Thailand
Actions
I’m Punsiri Boonyakiat, also known as Beat. I have accumulated over five years of experience in the tech industry, specializing in AI and data engineering. I also have the Google Cloud Certification as a Professional Data Engineer. One of my key projects at the company is integrating the LINE Messaging API with Vertex AI and developing chatbot solutions for retail. My work focuses on leveraging AI tools, such as Gemini Code Assist, to enhance data workflows and create innovative solutions in data-driven environments. I’m passionate about contributing to the community, not just as a speaker but also as a content creator. I’ve spoken at numerous events, including DevFest, LINE Dev Conference, Women Techmakers, Cloud AI Study Jam 2024 (ChaiyoGCP) in Thailand, and BuildwithAI Bangkok. You can find all my speaking engagements, writing, and projects on my website: https://punsiriboo.github.io/
Area of Expertise
Topics
Unlocking the Power of AI Agents in ADK with MCP Toolset
As enterprises move toward Agentic AI, the ability to seamlessly integrate conversational agents with real-world data and systems becomes essential. Google’s Agent Development Kit (ADK) provides a powerful foundation for building intelligent agents, while the Model Context Protocol (MCP) opens the door to a wide range of interoperable tools.
This session explores how developers can unlock the power of AI agents by combining ADK with the MCP Toolset. Using real-world scenarios, we will demonstrate how MCP bridges multiple domains—business intelligence (Looker), communication channels (LINE), and enterprise databases—into a single agent-driven workflow. By exposing these tools in a structured way, agents can deliver secure, reliable, and context-aware experiences to end users.
Skeleton MCP Toolset (for demonstration):
• Air BnB: for hotel search
• LINE Bot MCP Server: push messages, profile retrieval
• Looker Toolbox: query and dashboard access
• Database Toolbox: SQL queries
Building Open-Source AI Agents with Google Gemma and ADK
As AI agents move toward real-world adoption, developers are increasingly looking for open alternatives that provide greater control over cost, privacy, and deployment.
This session explores how to build AI agents using an open model approach with Google Gemma, integrated into the Agent Development Kit (ADK). Instead of relying on proprietary LLMs, we demonstrate how developers can leverage open models to design flexible and production-ready agent systems.
Building ADK Agents with Skills
As AI agents become more capable, managing context, expertise, and scalability becomes increasingly challenging. In this session, we'll explore how Google Agent Development Kit (ADK) Skills help developers build modular, efficient, and production-ready agents by loading capabilities only when needed. You'll learn how Skills reduce context overhead, improve agent performance, and enable reusable expertise across multiple agents. Through examples and live demonstrations, we'll cover Skill design patterns and integration with external tools to build an AI agent.
Ref: https://developers.googleblog.com/developers-guide-to-building-adk-agents-with-skills/
[GDG DevFest Bangkok 2024] - The Future of Retail E-commerce with Gemini AI
The talk showcases how Vertex AI and Gemini revolutionize retail e-commerce using Retrieval-Augmented Generation (RAG) to enhance product information and create a seamless shopping experience. The LINE Chatbot features three key functions: general product inquiries (Vertex AI Agent Builder), text-based product search (Vertex AI Search), and image-based product search (Gemini Image Understanding and Vertex AI Search)
[LINE Dev Conference 2024] - The Future of Retail e-commerce with Gemini AI
The talk showcases how Vertex AI and Gemini revolutionize retail e-commerce using Retrieval-Augmented Generation (RAG) to enhance product information and create a seamless shopping experience. The LINE Chatbot features three key functions: general product inquiries (Vertex AI Agent Builder), text-based product search (Vertex AI Search), and image-based product search (Gemini Image Understanding and Vertex AI Search),
Technologies: Vertex AI Search, Vertex AI Agent Builder, Gemini
[Chaiyo GCP - Online Code Along Session] - Gemini for Data Scientist [Running Codelab]
The lab session at the Chaiyo GCP Event in Thailand focuses on using Gemini in BigQuery as an assistant to build a K-means clustering model. As a speaker, I guided participants through segmenting customers based on their order behaviors by writing prompts for Data Scientists’s workflow. Technologies: Gemini
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top