Breaking Through a Career Bottleneck: From API Testing Stagnation to Go‑Powered Performance Engineering

Today I saw a question on a Q&A platform: How to break through a career bottleneck?

I will share my experience from last year to answer.

Last year I felt I had hit a dead end in interface testing; both automation and performance testing were routine, leaving me with only repetitive "brick‑moving" work and no challenge. I experimented with various performance scenarios such as full‑chain load testing, branch testing, geometric scaling, and improving data precision, but each effort lasted only a week or two before I moved to another angle. I briefly tried static code scanning, which did not yield good results, though I learned many fragmented Java concepts.

The situation changed after I switched to a new job where I focused exclusively on performance testing, encountering higher‑scale performance scenarios, tackling one challenge after another, and experiencing rapid progress comparable to a wild horse galloping across a prairie.

One Level, One World

In plain terms, a difference of an order of magnitude creates a different world: a single machine handling 1k QPS versus 10k QPS requires fundamentally different architectural solutions. Although performance‑testing engineers may use the same tech stack, the difficulties and test cases differ so much that the work can be considered two parallel professions.

As a book suggests, set strategies aimed at ten‑fold goals.

When we encounter a bottleneck, it usually means the current track offers little further growth, so we need to broaden or switch tracks. Looking back, I realize I stayed too long in my comfort zone and almost forgot the courage to pioneer.

Give yourself a ten‑fold goal, and the bottleneck becomes a stepping stone.

One level, one world; one multiple, one step.

One Go, One Universe

I mainly use Java and Groovy; now I can replace Java with Groovy wherever possible, and I once wrote Python for a short period. I used to say that mastering Java solves most work problems, and if something remains unsolvable, I would deepen my Java knowledge.

People grow through continual self‑questioning. This year I learned Go, which opened up a new world for me.

While the Java‑Groovy combo remains powerful, Go brings more fun and efficiency. It offers excellent testing frameworks such as goreplay and K6, and reading their source code inspired new ideas for the FunTester framework. I eagerly borrowed those ideas, re‑implemented some concepts in Java, expanding my perspective and providing fresh material for bragging.

Have Fun ~ Tester!

FunTester 2021 Summary

2022 Annual Plan Template

"Programmers Eliminated at 35" Already 22

Selenium Automation JUnit Parameterization Practice

Implementation of QPS Sampler in Performance Test Framework

Testing Non‑Fixed Probability Algorithm P=p(1+0.1*N)

Mobile Test Engineer Career

Groovy Hot‑Update Java Practice

Java Thread‑Safety ReentrantLock

Interface Test Code Coverage (Jacoco) Solution Sharing

Selenium Python Tips (Part 3)

Colored Console Output