[Hiring] AI Research & Evaluation Analyst @e2f

Role Description

We are looking for highly analytical professionals to support research and evaluation work for advanced AI and large language model (LLM) projects. This work requires more than traditional data annotation. Contributors must be able to understand complex information, evaluate factual claims and AI-generated content, apply sound judgment, and clearly explain their decisions in English.

Depending on project needs and demonstrated skills, contributors may support:

• Research and rubric development
• Rubric-based content evaluation

Previous AI or data annotation experience is helpful but not required. Strong research ability, analytical reasoning, attention to detail, and exceptional English comprehension and writing skills are more important.

Qualifications

• Fluent to near-native professional English proficiency (C1/C2 level or equivalent).
• Exceptional English reading comprehension, including the ability to understand nuance, qualifiers, ambiguity, and complex written claims.
• Strong analytical and critical-thinking skills.
• Ability to distinguish factual evidence from assumptions, opinions, and unsupported claims.
• Ability to produce clear, precise, grammatically strong written explanations.
• Strong attention to detail.
• Ability to follow complex written guidelines and apply them consistently.
• Ability to work independently and make evidence-based judgments.
• Strong digital literacy and online research skills.
• Ability to learn unfamiliar subject matter quickly.

Requirements

• Advanced web research and source-verification skills.
• Ability to identify authoritative sources appropriate to a specific claim.
• Ability to synthesize information from multiple or conflicting sources.
• Ability to translate complex research findings into clear evaluation criteria.
• Strong structured writing and authoring skills.
• Ability to anticipate ambiguity and create criteria that can be consistently interpreted by other evaluators.

Benefits

• This is a remote, project-based opportunity supporting AI research and evaluation projects.
• Project availability, workload, duration, and schedules may vary based on client requirements.
• Successful qualification does not guarantee a specific volume or duration of work.
• Project-specific requirements, rates, schedules, and guidelines will be communicated before assignment.

Preferred Experience

• Experience in one or more of the following is beneficial:
• Fact-checking or professional source verification
• Research analysis
• Evaluation or quality review
• Rubric or guideline development
• AI/LLM evaluation or training
• Investigative research
• Content quality or Trust & Safety
• Academic, legal, policy, or market research
• Working with primary sources, government databases, academic publications, corporate records, or other authoritative sources
• Previous AI experience is not required if you can demonstrate strong research, analytical, and evaluation capabilities.

Assessment Process

• Qualified applicants may be invited to complete a practical online assessment.
• The assessment may evaluate:
• English reading comprehension and written communication
• Analytical reasoning
• Fact-checking and source verification
• Source-quality judgment
• Evaluation of AI-generated content
• Ability to follow and consistently apply detailed criteria
• Ability to provide clear and concise rationales
• Candidates being considered for research and rubric-development assignments may complete additional exercises assessing research depth, synthesis, and rubric authoring.

What Success Looks Like

• Successful contributors don't simply identify whether something appears right or wrong. They can understand why, identify the evidence supporting that conclusion, and communicate their reasoning clearly.
• For research-oriented work, this means being able to research a topic and help define what a high-quality evaluation should measure.
• For evaluation-oriented work, this means being able to apply established criteria accurately, consistently, and efficiently while providing a concise rationale for the decision.