Full-Stack Software Engineer: Testing & Evaluation

What you'll do

You build the tasks that other engineers, and AI models, are tested on. The ad asks for technically substantial software projects that look like real engineering environments, structured as development challenges with clear documentation. You write deterministic test suites to check correctness and reproducibility, and you containerize each task with Docker so it runs the same everywhere.

In practice this is closer to building a benchmark than building a product: the test suite and the container matter as much as the code.

Who fits

• Proficiency in at least one modern language (JavaScript/TypeScript, Java, Go, Rust, C/C++ and similar)
• Experience architecting or maintaining multi-file projects
• Hands-on Docker or similar container experience; the ad calls it essential
• Strong technical writing and documentation
• Preferred: full-stack or end-to-end application work, and exposure to testing frameworks, build tools or code-quality workflows

What it pays

The board shows $75/hr, with no range in the ad text. The listing does not publish its payment terms or weekly hours.

Worth knowing

Good:

• Real engineering work with a clear output (a working, tested, reproducible project)
• Language-agnostic; you choose a stack you know
• Builds skills in test design and reproducible environments

Less good:

• Deterministic tests are harder than they sound; flaky tests are the common failure
• Payment terms and hours are not stated
• Open since December 2025, so check that it is still taking applicants

About this listing

Read from the AfterQuery Experts board on 18 October 2026 (posted 1 December 2025, still open). AfterQuery's help centre describes an unpaid, one-attempt assessment and ID verification for its expert pools. See AfterQuery.