Tom Simmons-AI Quality Analyst | LLM Evaluation Specialist

Worked full-time across commercial AI training projects evaluating proprietary Large Language Models for quality, factual accuracy, reasoning, safety and instruction following.

Highlights

• Selected for projects requiring native British English language expertise.
• Trusted to review and assess the work of fellow annotators as part of project quality assurance.
• Contributed across a wide range of AI evaluation projects covering reasoning, creative writing, factuality, research, safety, search quality, summarisation and mathematical reasoning.
• Worked successfully in fully remote, asynchronous environments requiring self-management, consistency and rapid adaptation to evolving project guidelines.

Responsibilities

• Evaluated AI-generated responses for factual accuracy, reasoning quality, clarity, instruction following and overall usefulness.
• Performed response ranking, preference ranking and prompt evaluation across a broad range of subject areas.
• Conducted independent research and fact-checking to verify AI-generated content.
• Evaluated outputs for safety, harmfulness, bias and hallucinations using detailed annotation guidelines.
• Produced structured written feedback to improve model quality, consistency and reliability.
• Applied complex evaluation guidelines consistently across multiple concurrent projects while maintaining high standards of accuracy.
• Collaborated effectively using Slack, Google Workspace and other online collaboration tools.

• Investigated complex tenancy, legal and anti-social behaviour cases using evidence-based decision making.
• Analysed conflicting information, assessed credibility and reached balanced conclusions.
• Prepared detailed reports and documentation for legal proceedings.
• Interviewed witnesses and gathered evidence from multiple sources.
• Managed sensitive personal information in accordance with legal and organisational requirements.