Francis Wanyoike Muiruri
I’m Francis Wanyoike Muiruri, an AI Quality Analyst and LLM Evaluation Specialist based in Nairobi (remote).
About
I’m Francis Wanyoike Muiruri, an AI Quality Analyst and LLM Evaluation Specialist based in Nairobi (remote). I evaluate AI-generated responses and reasoning-heavy outputs against clear quality standards—focusing on accuracy, relevance, clarity, coherence, completeness, and strong instruction following. My work often involves comparing multiple model outputs side-by-side to identify subtle differences, unsupported claims, logical gaps, and other issues that affect user usefulness and reliability.
I also support prompt and conversation testing by designing evaluation scenarios that measure comprehension, grounding/factuality, and edge-case behavior in multi-turn interactions. Alongside evaluation, I’ve contributed to structured dataset work through annotation, guideline-based scoring, and taxonomy adherence, and I’ve produced clear, evidence-based feedback to help teams improve both responses and evaluation frameworks. I’m fluent professionally in English and comfortable working independently on task-based, remote workflows with high consistency and attention to detail.
- 8 years' experience
- Joined September 2026
Languages
- EnglishFluent
- SwahiliFluent
- PolishAdvanced
Industry experience
- Education
- Professional Services
- Media & Entertainment
- Software & Internet
Work experience· 4 roles
Supported advanced AI training and evaluation workflows by assessing AI-generated responses and reasoning-heavy outputs against structured criteria (accuracy, relevance, clarity, coherence, reasoning quality, completeness, and instruction adherence). Performed comparative side-by-side analysis across multiple model outputs, ranking responses and explaining decisions with concise, evidence-based rationales. Designed and evaluated prompts and evaluation scenarios to test reasoning, comprehension, instruction following, and conversational quality, including multi-turn settings and edge cases. Identified factual inconsistencies, unsupported claims, contextual errors, ambiguities, and logical gaps, and contributed to refinement of evaluation guidelines used in AI training workflows.
Conducted structured evaluation of digital and educational content using predefined quality criteria and scoring frameworks. Assessed clarity, coherence, relevance, usability, and comprehension, comparing alternative content versions to determine relative quality and effectiveness. Identified ambiguity, inconsistencies, missing information, weak reasoning, and unclear communication, and evaluated how well content addressed user needs in a natural, understandable way. Supported multilingual evaluation workflows while maintaining consistent quality standards and documented findings with actionable feedback for improvements.
Supported AI training datasets through structured annotation, response evaluation, and quality assessment. Evaluated written outputs and model responses using guideline-based scoring and classification criteria, checking linguistic accuracy, logical consistency, relevance, clarity, and completeness. Performed comparative evaluations of multiple responses and ranked outputs according to predefined quality standards, providing documented reasoning for evaluation decisions. Identified errors, inconsistencies, and ambiguous outputs to be corrected as part of quality assurance for large-scale training projects.
Education
Hire Francis Wanyoike Muiruri today
Post a job and invite Francis, or contact them directly. No hidden commission.