JobHabor
Menu

AI Evaluations Researcher

Mondo · San Francisco, CA · Hybrid

Strong hiring record · median 23 days open

100100

Hiring signal

Actively hiring

Posted today · Never reposted · Checked today · How we score

First seen
19 November 2026
Reposted
Not since September 2026
Last checked
Today · still on the employer's board
All reasons

In its favour

  • Posted today
  • Still listed on the employer's own board
  • Publishes a salary range

Did you apply?

AI Evaluations Researcher at Mondo. We’ll keep it in your applied jobs, so you can see it later and be warned before applying to the same employer twice.

AI Evaluations Researcher

Location

San Francisco, CA (Onsite 3-4 days per week)

Schedule

Full-time, 40 hours per week

Contract Length

1+ Year Contract

Compensation

$82 - $92 / hour W2

About the Role

We're seeking an AI Evaluations Researcher to help analyze, interpret, and communicate the performance of frontier AI models. This role sits at the intersection of AI research, benchmarking, technical analysis, data visualization, and thought leadership.

The ideal candidate is deeply engaged with the AI ecosystem, understands how model evaluations and benchmarks work, and can transform complex findings into clear, accurate, and compelling research content. You'll work with benchmark results, AI performance data, economic and workforce datasets, and emerging model releases to produce insights that resonate with both technical and business audiences.

This is an opportunity to contribute to discussions shaping the future of AI evaluation while building research that helps organizations understand what today's models can and cannot do.

Key Responsibilities

Research & Analysis

  • Analyze AI benchmark results and model evaluation data
  • Assess the performance of frontier AI models across a range of tasks and use cases
  • Conduct research on emerging trends in AI capabilities, evaluation methodologies, and model benchmarking
  • Track developments across the AI ecosystem and competitive landscape
  • Perform business and market analyses related to model performance, costs, and benchmark results

Technical Writing & Thought Leadership

  • Author research reports, technical briefs, and analytical summaries explaining benchmark findings
  • Create data-driven content that communicates technical concepts accurately and clearly
  • Develop concise, high-quality commentary on model releases and AI industry developments
  • Produce technical and research-focused content for web and social media channels
  • Translate complex evaluation results into practical business and real-world implications

Data Visualization & Reporting

  • Build charts, graphs, and visualizations that communicate research findings effectively
  • Compare model performance across quality, cost, and task-based metrics
  • Present benchmark outcomes with appropriate context, limitations, and methodology considerations
  • Maintain reporting frameworks and standards for research publications

Industry & Economic Context

  • Connect benchmark results to real-world tasks, job functions, labor markets, and economic outcomes
  • Leverage workforce, occupational, and wage datasets to contextualize AI performance findings
  • Communicate the practical impact of evolving AI capabilities beyond benchmark scores alone

Required Qualifications

  • 2-5 years of experience in AI research, evaluations, benchmarking, policy research, analytics, or a related field
  • Deep understanding of AI evaluations, benchmark frameworks, and model assessment methodologies
  • Ability to explain the strengths, weaknesses, and limitations of benchmarks and evaluation systems
  • Strong familiarity with AI model development, performance testing, and AI data ecosystems
  • Proficiency with Python or R for research, analysis, and visualization
  • Excellent written communication skills and the ability to produce polished research content
  • Strong data visualization and analytical capabilities
  • Demonstrated ability to make accurate, evidence-based claims and avoid overstating findings
  • High level of ownership, responsiveness, and ability to work independently

Preferred Qualifications

  • Experience building, managing, or operating AI benchmark or evaluation systems
  • Familiarity with LLM evaluations, task-based assessments, and model benchmarking frameworks
  • Experience working with workforce, occupational, labor market, or economic data
  • Published research, industry reports, working papers, newsletters, blogs, or Substack articles
  • Strong short-form writing skills, including social media and technical commentary
  • Active involvement in the AI community through writing, content creation, research, or public discussion on X

More jobs at Mondo

All 42

Similar roles