Research & Evaluation Project

Revelo; United States; Remote; USD 55,000 - 110,000; Full time

Job summary

Revelo seeks a Research Specialist in LLM Training to perform qualitative and structured evaluations that improve AI performance. You will review prompts for clarity and domain alignment, assess multiple AI responses for accuracy and depth, and compare options across factuality and reliability.

You will provide actionable feedback to curb hallucinations and guide model improvements. This freelance, flexible-hours role emphasizes technical writing and precise analysis to help refine AI systems

Qualifications

• Reviewing prompts for clarity, relevance, and domain alignment.
• Analyzing AI-generated responses for accuracy, depth, and formatting.
• Comparing responses across dimensions like factuality and source reliability.
• Providing actionable feedback to flag hallucinations and improve model behavior.

Responsibilities

• Prompt Analysis: Review user prompts for clarity and relevance.
• Response Evaluation: Assess AI outputs for accuracy and depth of research.
• Comparative Assessment: Identify preferred responses and explain choices concisely.
• Feedback & Edge-Case Identification: Flag hallucinations and suggest improvements.

Skills

Research Analysis

Fact-Checking

Critical Thinking

Technical Writing

Job description

Main Activities

As a Research Specialist in LLM Training, you will engage in qualitative and structured evaluations to enhance AI performance. Your key activities will include:

• Prompt Analysis:Reviewing user prompts to assess clarity, relevance, research context, and domain alignment.
• Response Evaluation:Analyzing multiple AI-generated responses to evaluate accuracy, depth of research, formatting, and adherence to instructions.
• Comparative Assessment:Determining the preferred AI response based on established dimensions (such as factuality, reasoning quality, and source reliability) and writing concise explanations for your choices.
• Feedback & Edge-Case Identification:Providing actionable feedback and flagging hallucinated or inaccurate research insights to help refine future model iterations.

Skills

Research Analysis, Fact-Checking, Critical Thinking, Technical Writing

Duration

Continuous / Fast-Moving

New to LLM Training?These are projects where you use your technical expertise to help improve AI models. You'll work on tasks like evaluating code, testing AI-generated solutions, and providing feedback that directly teaches AI systems to become better at programming. This is freelance work with flexible hours, so you can work on your own schedule. Highly recommended read here: https://bit.ly/llm-training-projects.