AI Benchmark Engineer | Native Language Specialist - Spanish (Argentina) - Remote

LILT (Production); Argentina; À distance; ARS 41 729 000 - 83 458 000; Plein temps

Résumé du poste

LILT Production in Argentina is seeking experienced software engineers to design, build, and validate multilingual benchmarks. You will create high-signal tasks in native languages, test models on non-English data, and ensure the assets stay in the target language.

This remote, freelance role emphasizes rigorous QA, deterministic verifiers, and calibration across model tiers, with emphasis on grammar accuracy and locale handling.

Qualifications

• Experience: 1+ years of industry experience in software engineering.
• Background: Proven track record at leading technology companies and/or graduation from top-tier engineering universities.
• Language: Native or near-native fluency, with a deep understanding of its gramma

Responsabilités

• Task Engineering: Evaluating Coding Agents.
• Asset Creation: Build realistic task environments using datasets and files in your native language. Crucially, these assets must remain in the target language to genuinely measure multilingual handling.
• Prompting & Translation: finding failure points where AI does not work, in your native language Implementation & Verification: Support the development of robust solutions (reference implementations) and write highly reliable, deterministic verifier scripts (using rubric-based judging only when strictly necessary).
• Calibration & Execution: Analyze execution logs and calibrate task difficulty (Easy to Very Hard) using standard Terminal-Bench run configurations against various model tiers (Haiku, Sonnet, Opus).
• Quality Assurance: Participate in a rigorous, 4-layer human quality control process (creation, human review, calibration review, and audit) alongside automated LLM-based checks to ensure fairness, grammatical accuracy, and benchmark integrity.

Connaissances

Software engineering

Native fluency

Description du poste

About The Opportunity

We are building a rigorous, verifiable evaluation suite of Terminal-Bench tasks designed to test the limits of large language models on multilingual software challenges. Our goal is to measure multilingual robustness across prompt language effects, non-English data processing, and complex locale/encoding edge cases in terminal workflows. We are seeking experienced native-speaking software engineers to design, build, and validate these benchmarks. You will create high-signal, high-quality tasks that genuinely test a model's ability to handle multilingual environments without relying on English translation crutches. Note this is a remote, freelance opportunity

What You’ll Deliver

• Task Engineering: Evaluating Coding Agents.
• Asset Creation: Build realistic task environments using datasets and files in your native language. Crucially, these assets must remain in the target language to genuinely measure multilingual handling.
• Prompting & Translation: finding failure points where AI does not work, in your native language Implementation & Verification: Support the development of robust solutions (reference implementations) and write highly reliable, deterministic verifier scripts (using rubric-based judging only when strictly necessary).
• Calibration & Execution: Analyze execution logs and calibrate task difficulty (Easy to Very Hard) using standard Terminal-Bench run configurations against various model tiers (Haiku, Sonnet, Opus).
• Quality Assurance: Participate in a rigorous, 4-layer human quality control process (creation, human review, calibration review, and audit) alongside automated LLM-based checks to ensure fairness, grammatical accuracy, and benchmark integrity.

Qualifications

• Experience: 1+ years of industry experience in software engineering.
• Background: Proven track record at leading technology companies and/or graduation from top-tier engineering universities.
• Language: Native or near-native fluency, with a deep understanding of its gramma