//assets.cakeresume.com/assets/contentarea/contentarea-2d64eca8dae2a1ffb31e0fb1df887bbad072c7c19734d7d30fa14884bbcc299a.css

1.帶領團隊透過Docker建構大數據分析平台，並且達到99%可用性與低於10秒的數據查詢時間。

2.規劃資料庫無縫遷移計畫(MariaDB>PostgreSQL+Timescale)，解決資料量達到80M時遭遇的查詢效能瓶頸。

3.建構環境領域專用的GIS和圖表分析工具，協助客戶成為公共民生物聯網最大的空品計畫承包商，裝置維運數量擴大至3500台。

4.整合Clickhouse解決方案，緩解資料量達到3B時遭遇的查詢效能瓶頸。

5.重構微服務為Pub/Sub架構，降低60%的雲端計算費用並提供無縫整合開源專案的彈性。

6.運用AI模型開發環境異常預警，輔助台中市政府智慧稽查，更獲得"Smart 50 Awards"。

Air IoT Analytics Platform

Big data, AI, Blockchain

蔡智鈞

I like to use software techniques yo solve real problems. Recently, I am mainly focusing on LLM and zero knowledge technologies.

* Demonstrated expertise in crafting, testing, and deploying robust smart contracts utilizing the Foundry framework within the Ethereum ecosystem.
* Gained comprehensive insights into the technical structures and operational mechanisms of leading DeFi protocols including Compound, Uniswap, AAVE, and Balancer.
* Conducted extensive analysis of smart contract vulnerabilities such as re-entry attacks, denial-of-service issues, and oracle manipulation.
* Developed a DeFi protocol which allows traders to manage investor’s funds while ensuring funds safety.

AppWorks School

國立暨南國際大學(National Chi Nan University)

Principle Engineer

Build a source code scanning pipeline with CodeQL on Jenkins for 150+ projects.
Build a packages scanning pipeline with Yarn on Jenkins for 30+ projects.
Implement cache mechanism to archive 40% performance optimization for Jenkins pipeline.

台灣台北市

# Air quality IoT analytics system (2019 - 2023)
1. Led a team in building an IoT device management and data analytics system using microservice architecture with Docker. Achieved 99% SLA and under 10 seconds data query time.
2. Designed and executed a seamless migration plan from MariaDB to PostgreSQL with Timescale extension, addressing a query performance bottleneck caused by 80 million records.
3. Developed a stable and practical analysis platform that facilitated the customer's expansion from 500 devices to 3,500, establishing them as the leading consulting company in Taiwan for Civil IoT initiatives in collaboration with local governments.
4. Integrated ClickHouse to overcome query performance bottlenecks as the data volume reached 3 billion records.
5. Transitioning from an API-based approach to Pub/Sub architecture. This transformation resulted in a 60% reduction in computing costs.
6. Implemented an AI model for PM2.5 data calibration and anomaly detection, enabling governments to accurately identify the sources of pollution. This solution contributed to the organization winning a prestigious "Smart 50 Award."

# Digital Lottery System (October 2021 - February 2022)
1. Spearheaded the development of a digital lottery system aimed at revitalizing the economy of Puli post-epidemic. This system garnered an impressive participation of over 6,000 individuals and collaboration with 150+ stores, generating 3.5 million economic activities.

# Agriculture Irrigation and Monitoring System (2018 - 2023)
1. Developed a web visualization system using React.js and MongoDB. This approach resulted in an impressive 20% reduction in daily workload.

# Air quality real-time monitoring map (2015-2022)
1. Utilized the ASP.NET MVC framework for building a low-cost PM2.5 sensing platform for the Environment Protection Administration Executive Yuan R.O.C.
2. Integrated Google Maps API to display GIS base data, and leveraged Apache EChart to present historical data.
3. Developed a Messenger Bot to provide convenient data querying services.

Share