ELL3N HSU

ELL3N HSU

Python engineering · Test platforms · Payments infrastructure

Profile

Python engineer with 7+ years of experience, including 5+ years on-site at WeChat Pay building test infrastructure for money-movement products. I designed the shared merchant fund test SDK, now reused by 23 business classes across 5 product areas, and was a core contributor to an api testability library, taking two core payment lines from hand-written scripts to reusable APIs on a unified SDK. Over the past two years I have brought LLMs and AI agents into test production with a staged spec-to-test pipeline, delivering new-product acceptance tests about 7× faster. English-major background; I work directly from English primary sources.

Experience

Senior Python Developer → Software Engineer in Test

— Present

Vendor · On-site at Tencent WeChat Pay

Own test infrastructure for WeChat Pay money-movement products (deduction pay, merchant payouts, recharge, withdrawal). On-site since Mar 2021, growing from test author to test SDK designer and explorer of the AI testing workflow.

  • 1,500+ test files I delivered
  • 266 APIs in the shared library
  • 23 classes on the SDK
  • 7× faster acceptance
  1. 2025–now
    AI testing workflow, 0→1

    Built a staged LLM pipeline from API specs to executable tests, paired with an AI-agent delivery flow and human review gates. A new product's acceptance suite shipped 200+ test files in about 4 weeks, roughly 7× faster.

  2. 2024–2025
    Test SDK architecture

    Led a two-layer refactor of the merchant fund test SDK (transport vs. business semantics), unifying signing, return-code retries, and version routing while keeping 20+ existing subclasses backward compatible. 23 business classes across 5 product areas now build on it.

  3. 2022–2023
    Testability API platform

    Pushed protocol assembly and signing down into parameterized APIs so tests only state scenarios and assertions. As a core contributor, helped grow the library to 266 interfaces; contract-signing APIs alone support 1,800+ test files.

  4. 2021
    Scripts → framework

    Drove the migration of legacy recurring-deduction tests onto the new framework: extracted shared protocol and data setup, then migrated in batches until every case passed, laying the ground for API-based testing.

Earlier

  • 2019.03–2020.07
    Python backend developer

    Built web backend APIs and data-processing jobs; grounding in Python engineering, databases, and API design.

  • 2015.07–2018.02
    Overseas technical support · Tenda · Reolink

    Diagnosed router and networking issues for overseas customers and channel partners, working in English. Studied Python full-time from Mar 2018 and moved into engineering in 2019.

Personal code projects

Payments test APIs and shared SDK

Share request behavior across business APIs while handling protocol changes, existing callers, and retry conditions.

XAT · Python API testing

Designed and iterated XAT, a Python API testing tool using unittest with process concurrency, method deadlines, logs, and JSON reports.

AI testing workflow

Connect staged LLM generation, human review, and execution checks to turn generated code into useful test cases.

Skills & study

AI testing engineering

LLM application development, staged and versioned prompt engineering, AI agent skill and memory design, tool-style API design for function calling and MCP, human review gates, and evaluation regression for generated output.

Test platform engineering

Layered test SDK architecture, testability API design, API and contract testing, mock injection, declarative retries, and flaky-test root-cause analysis; Python (including 3.6-compatible code), unittest, Protobuf/RPC, REST, Jinja2, and Git.

Payments domain

Money-movement flows including deduction pay, merchant payouts, recharge, and withdrawal; fund state machines and terminal-state consistency, idempotency and duplicate requests, async settlement and callback timing etc

Project delivery

Requirements and metric design, AI-assisted development, Git reviews, CI regression and acceptance, Docker Compose, and Nginx deployment.

Research focus

LLM serving: prefill and decode, KV cache, prefix caching, continuous batching and paged KV, load-testing metrics (TTFT, TPOT, goodput), and reproducible experiment methods.

Languages

Native Chinese; English-major degree and 3 years of front-line technical support in English. I read papers and primary docs directly and write design documents in English.

Education

Chang’an University

Bachelor’s degree in English · 2011–2015

Contact