Backend Engineer for AI Evaluation Tasks (Remote)

Weekday; United States; Remote; USD 69,000 - 138,000; Part time

Job summary

Weekday is seeking experienced Backend Software Engineers to craft repository-based coding tasks that test AI-model debugging, backend feature work, and performance optimization. This fully remote, project-based role pays per approved task and requires hands-on experience in Python, Java, Rust, Go, C++, or TypeScript.

You will build reproducible environments, write deterministic verifiers, and develop reference solutions while contributing to challenging RL environments used to evaluate AI

Qualifications

• Strong backend development experience in Python, Java, Rust, Go, C++, or TypeScript.
• Solid understanding of data structures, algorithms, and software architecture.
• Ability to debug complex backend systems and implement fixes.
• Experience building backend features and improving performance.
• Design reproducible test environments and deterministic verification mechanisms.

Responsibilities

• Design realistic backend engineering tasks using repositories and codebases you know.
• Develop problems involving bug fixes, feature implementation, API development, refactoring, and performance tuning.
• Build reproducible environments for executing and evaluating tasks.
• Create deterministic verifiers that handle multiple valid implementations.
• Develop reference solutions showing effective solving approaches.
• Evaluate tasks for clarity, relevance, and verifiability.
• Document task requirements and verification criteria clearly.

Skills

Python

Java

Rust

Go

C++

TypeScript

Job description

Weekday is seeking experienced Backend Software Engineers to craft repository-based coding tasks that test AI-model debugging, backend feature work, and performance optimization. This fully remote, project-based role pays per approved task and requires hands-on experience in Python, Java, Rust, Go, C++, or TypeScript.

You will build reproducible environments, write deterministic verifiers, and develop reference solutions while contributing to challenging RL environments used to evaluate AI