1. Team Members

  • Me (Contributed 90%)
  • Another Engineer (Contributed 10%)

2. Company

Shenzhen Qutui Science & Technology Co., Ltd.

3. Technologies

3.1 Languages & Frameworks

  • Node.js (TypeScript) (80%)
  • Go (20%)
  • Egg.js
  • go-kratos

3.2 Databases & Caching

  • MongoDB
  • Redis

3.3 Messaging & Queueing

  • RabbitMQ

3.4 Infrastructure

  • Docker, Kubenaties
  • Tencent, Alibaba Cloud

4. Introduction

Video Rendering Service is an internal backend core service that supports all the company’s video generation products.

It integrated 2 video rendering engines to implement many kinds video generation, providing REST API to interact with other backend projects. The architecture is as follows.

5. My Role

I led the development and acted as a bridge to coordinate with multiple teams, including 5 engineer teams (11 members). And sometimes deal with feedback from other non-engineers, such as project manager, product manger, video designers. The cross team work relationship as follows.

6. Biggest Challenge

As you can see from the diagram, I was facing a huge amount of communication work while also being the core developer of the Flash Cut project. Every day, I was constantly interrupted—whether it was debugging rendering failures, answering questions from different teams, or troubleshooting unexpected issues. I couldn’t even get 20 minutes of uninterrupted focus to code or debug properly. But the development deadline didn’t change. It was extremely overwhelming

How I overcome it

I made a few key changes:

  • Prioritization & Communication: I sat down with my team leader and project manager to clearly define what was urgent vs. what could wait. This helped set expectations and reduced unnecessary interruptions.

  • Logging Improvement: I improved the logging output, making it easier for other engineers to find the root cause of rendering issues without needing my immediate help.

  • Documentation: I created a comprehensive tech guide, including API usage, troubleshooting guide, and cross-team collaboration guide so that other teams could identify and solve common issues by themselves instead of pinging me for every small problem.

The Result
  • I regained control of my workflow and finally had time to focus on actual development.
  • The multiple team collaboration productivity improve obvious, reducing unnecessary communication and interrunption.
  • I got an A- performance review (1st of 14) in the back-end team in April 2022. (Every month, each team cloud only zero or one member get A- or A).
What I learned from this experience

  • Communication is the Key: Keeping open communication and aligning priorities with my manager and team reduced stress and improved efficiency.

  • Work Smarter, Not Just Harder: Building better logs and writing comprehensive documentation took extra effort upfront, but in the long run, it saved me (and everyone else) a ton of time.

  • Prioritize Everything: Not everything is urgent. By setting boundaries and prioritizing properly, I got more done instead of trying to handle everything immediately.

  • Balance is Key: There will always be urgent tasks, problems to solve, and unexpected issues. Instead of just working harder, I learned to step back, organize, and build better systems — making a difference for both me and coworks.