Flutter Developer — United States

About The Company

LaraBench is building tools and processes to help teams assess the quality, reliability, and real-world performance of artificial intelligence systems. We believe that useful AI evaluation requires clear criteria, thoughtfully designed tests, and careful human judgment. Our work brings together technology, research, and practical problem-solving to make model performance easier to understand and improve. At LaraBench, we value curiosity, accuracy, collaboration, and responsible innovation. We are looking for people who enjoy working through complex questions and can turn detailed observations into clear, actionable insights.

About The Role

As an AI Benchmarking Specialist, you will contribute to the design, execution, and analysis of evaluations for AI systems. You will review model responses, apply consistent evaluation criteria, identify patterns in performance, and document findings for internal teams and clients. The role combines analytical thinking with strong written communication and attention to detail. You will work with researchers, product professionals, and technical colleagues to ensure evaluations are relevant, repeatable, and useful for decision-making.

This position is well suited to someone who is comfortable navigating ambiguous questions, comparing multiple possible answers, and explaining the reasoning behind a judgment. You will receive guidance on LaraBench’s evaluation methods and have opportunities to help improve them as our projects evolve. Depending on project needs, assignments may include language, reasoning, instruction-following, factuality, or other dimensions of AI performance.

Qualifications

We welcome candidates with backgrounds in research, linguistics, data analysis, education, technology, communications, or other fields involving careful assessment and written work. A degree or equivalent practical experience is valued; relevant skills and demonstrated ability are equally important. You should be able to read complex instructions, apply guidelines consistently, and explain conclusions in clear, professional English. Strong attention to detail and sound judgment are essential.

Successful candidates are comfortable using online tools and spreadsheets, managing competing priorities, and asking thoughtful questions when requirements are unclear. Experience evaluating AI-generated content, conducting research, reviewing written material, or working with structured datasets is an advantage, but is not required. We value intellectual curiosity, discretion when handling project information, openness to feedback, and a commitment to fair, evidence-based evaluation.

Responsibilities

Review AI-generated outputs against project-specific criteria, examples, and quality standards. Assess response quality across relevant dimensions, record ratings and supporting rationale, and flag unclear, inconsistent, or potentially biased evaluation guidance. Conduct structured comparisons of model outputs and identify recurring strengths, limitations, and failure patterns.

Work with project leads and colleagues to clarify evaluation objectives, maintain accurate documentation, and meet agreed deadlines. Contribute feedback that can improve rubrics, workflows, and quality-control processes. Participate in calibration activities to support consistent judgments across evaluators, revise assessments when needed, and communicate observations in a constructive manner. Handle project materials responsibly and follow applicable privacy, security, and data-handling procedures.

Benefits

LaraBench offers the opportunity to contribute to meaningful work in a developing area of technology and to collaborate with colleagues who value careful thinking and practical impact. Team members receive clear project guidance, constructive feedback, and opportunities to build skills in AI evaluation and research methods. Compensation, work arrangements, and any additional benefits will be discussed during the hiring process and may vary based on role requirements, location, and employment status.

Equal Opportunity

LaraBench is committed to providing an inclusive and respectful workplace. We consider applicants without regard to legally protected characteristics and welcome candidates with varied experiences, perspectives, and career paths. If you need an accommodation during the application or interview process, please let us know so we can support your participation.