From Agile to AI: Challenges and Solutions for Enterprise Software Documentation

  • Presentation
  • Software Documentation
  • 12. November
  • 11:15 AM (CET) - 12:00 PM (CET)
  • C5.1
  • finished
  •  Sherrie Pan

    Sherrie Pan

    • Shanghai Yunzhou Technology Co.,Ltd.

Contents

In this presentation, I will share my experience leading a documentation team to efficiently and agilely develop and deliver documentation for a series of cloud software products.

I will introduce how we built a DITA-based technical documentation repository from scratch, established a continuous integration and continuous delivery (CI/CD) pipeline for documentation automation, and fine-tuned and deployed a large language model (LLM) to enhance documentation translation.

Takeaways

1. Implement an enterprise-ready CI/CD pipeline to reduce documentation delivery time by 50%.  

2. Apply LLM fine-tuning techniques to improve multilingual documentation accuracy.  

3. Balance agility with compliance in cloud software documentation workflows.  

Prior knowledge

1. Basic knowledge of Cloud Computing IaaS

2. Basic knowledge of Software DevOps and related toolchains

3. Basic knowledge of Structured Documentation Development and Compilation/Release

4. Basic knowledge of AI Infrastructure

5. Basic knowledge of LLM Model Deployment and Fine-tuning

Speaker

 Sherrie Pan

Sherrie Pan

  • Shanghai Yunzhou Technology Co.,Ltd.
Biography

Technical Documentation Manager at ZStack. Leads a team that built a DITA-based documentation system delivering tens of millions of characters for a series of enterprise software products. Currently leading the team's transition toward documentation intelligence and knowledge engineering. Many years of deep experience in cloud computing IaaS, with extensive expertise in software documentation development and management.