Open to opportunities

Yves User

@yvesavilla

Message

Freelance Python Developer and AI coding evaluator, comparing LLM code for production-ready PR-quality.

Brazil

Message

What I'm looking for

I’m looking to evaluate and improve AI-generated Python code with a PR-quality lens—focusing on correctness, maintainability, and test coverage—while building reproducible setups and running SWE benchmarks to generate reliable human preference data.

I’m a Freelance AI Coding and AI Model Comparison specialist with 3+ years evaluating LLM-generated code against production-readiness standards (PR quality). I focus on correctness, maintainability, test coverage, and engineering workflow, using PR-style review and rubric-based scoring to produce clear, structured findings.

I also implement and improve Python codebases with a strong test-first, coverage-aware mindset, including reproducible dev setups with Docker and clean, CI-friendly runs. From SWE-style benchmarks to side-by-side model comparisons, I design realistic tasks, run regression checks, and deliver human preference data—often in iterative, multi-agent/multi-trajectory “code review” workflows.

Experience

Work history, roles, and key accomplishments

Current

Freelance AI Coding Evaluator

Current

Freelance

Jan 2023 - Present (3 years 6 months)

Evaluated LLM-generated code on real Python repositories using PR-style quality and production-readiness standards, including readability, architecture fit, edge cases, security considerations, and code-style alignment. Ran side-by-side model comparisons and produced rubric-based human preference data, while building reproducible Docker-based dev setups and maintaining reliable pytest regression s

Style Review Side By Side Model Comparison Task Design Pass Fail Criteria Definition Code Reproducibility Docker Pinned Dependencies Pytest Coverage Improvement Git Code Review Clean PRs Regression Testing Benchmarking Error Handling Interface Design

Current

Freelance Python Software Eng

Current

Freelance

Jan 2018 - Present (8 years 6 months)

Implemented features and bug fixes across Python codebases with test-first and coverage-aware changes, emphasizing refactoring for maintainability. Created environment setup docs/scripts to make local and CI runs consistent and delivered production-focused patches with clean diffs.