Mohamed Radwan

AI Evaluation Specialist | Data Analyst | Research Consultant
Expert in evaluating AI-generated outputs, designing rubrics, and delivering structured analytical feedback for frontier model development. Proven track record in data analysis, market research, and building AI-assisted workflows.
AI Response Evaluation Rubric Design Data Analysis Market Research Prompt Engineering Python SQL Product Analytics
12+
AI Evaluation Projects
4
Years Analytics
500+
Rubrics Designed
98%
Feedback Accuracy
3
Industries Served

Experience

AI Evaluation Specialist — Freelance2024 — Present
  • Design and evaluate structured rubrics for AI model output assessment across generalist and domain-specific tasks, delivering 95%+ inter-rater reliability
  • Evaluate 200+ AI-generated responses per week, providing detailed written rationales identifying nuance, reasoning gaps, and factual accuracy
  • Develop custom evaluation frameworks for frontier AI labs, calibrating model performance against human expert baselines
  • Mentor 3 junior evaluators on rubric design and critical assessment methodology
Data Analyst — AIToolPick2025 — Present
  • Built and maintain a data-driven affiliate review platform serving 5K+ monthly visitors, using Python-based analytics to optimise content strategy
  • Developed automated market intelligence pipelines tracking competitor pricing, feature comparisons, and user sentiment across 50+ AI tools
  • Designed structured evaluation criteria for AI product reviews, improving content quality scores by 40%
Product & Research Analyst — ROKZAN2025 — Present
  • Lead data-driven product improvements for a B2B transportation management platform serving 500+ corporate users
  • Perform quantitative analysis of user behaviour data to identify optimisation opportunities, increasing route efficiency by 25%
  • Synthesise user research and feedback into actionable product recommendations presented to C-suite stakeholders
Research Assistant — Cardiff Metropolitan University2024 — 2025
  • Conducted structured literature reviews and data synthesis for cross-disciplinary research projects
  • Analysed quantitative datasets using Python (pandas, numpy) to identify statistically significant patterns
  • Co-authored research summaries requiring precise analytical writing and evidence-based reasoning

Key Projects

AI Response Evaluation Framework

A comprehensive rubric-based evaluation framework for assessing LLM outputs across accuracy, reasoning, coherence, and safety dimensions. Used to evaluate 500+ model responses with consistent scoring.

Rubric DesignEvaluationPrompt Engineering
View on GitHub →

Data Analytics MCP Pipeline

Automated data analysis pipeline using Model Context Protocol for structured data extraction, transformation, and insight generation. Processes 10K+ records per run.

PythonData AnalysisAutomation
View on GitHub →

Affiliate Intelligence Dashboard

Market intelligence platform tracking 50+ AI SaaS products with automated competitor analysis, pricing monitoring, and sentiment aggregation from 500+ review sources.

Market ResearchAnalyticsWeb Scraping
View on GitHub →

Agentic Workflow Suite

Designed and implemented multi-agent orchestration workflows for automated task delegation, quality control, and structured output generation across diverse analytical contexts.

AI AgentsWorkflow DesignAutomation
View on GitHub →

Core Competencies

AI Evaluation

Response evaluation, rubric design, prompt engineering, red teaming, bias detection, factuality assessment, inter-rater reliability

Data & Analytics

Python (pandas, numpy, scipy), SQL, statistical analysis, data visualisation, A/B testing, trend analysis, quantitative research

Research

Structured literature review, qualitative analysis, market intelligence, competitive analysis, user research synthesis, technical writing

Tools & Platforms

Git, GitHub Actions, SQL databases, Google Analytics, Looker Studio, API integration, CI/CD pipelines, Linux, cloud platforms

Education

BSc Computing — Cardiff Metropolitan University2023 — 2026
  • Specialising in data analysis, AI systems, and human-computer interaction
  • Dissertation: AI-Assisted Decision Making — Evaluating LLM Output Reliability
  • Dean's List recognition for academic excellence

Email: mohamed.origami@gmail.com · Cardiff, UK

Available for AI evaluation, data analysis, and research consulting projects