Ondieki O.

Key Skills

Software

Micro1

Mindrift

Appen

OneForma

Remotasks

Scale AI

Toloka

Top Subject Matter

Large Language Model (LLM) evaluation and safety alignment

LLM prompt engineering and evaluation

Legal Services & Contract Review

Top Data Types

Text

Document

Top Task Types

Prompt + Response Writing (SFT)

Freelancer Overview

AI Model Evaluation & Data Annotation Specialist Contract / Project (2025 – Present). Brings 2+ years of professional experience across legal operations, contract review, compliance, and structured analysis. Core strengths include Internal and Proprietary Tooling. Education includes Bachelor of Science, University of Nairobi (2025). AI-training focus includes data types such as Text and labeling workflows including Evaluation, Rating, and Prompt + Response Writing (SFT).

Labeling Experience

AI Model Evaluation & Data Annotation Specialist Contract / Project (2025 – Present)

Text

Evaluated and graded 500+ LLM-generated responses for logical reasoning, factual accuracy, mathematical correctness, and safety guideline adherence. Performed structured review of outputs to ensure compliance with complex constraints and to identify issues affecting alignment and trust. Used scoring and rubric-based judgment to improve dataset quality and downstream model behavior. • Assessed reasoning correctness, factuality, and mathematical validity • Checked adherence to complex safety guidelines • Identified and reported subtle contradictions and semantic errors • Supported training alignment through iterative feedback

2025 - Present

AI tutoring/training-related prompt engineering within AI Model Evaluation project (2025 – Present)

TextPrompt + Response Writing (SFT)

Developed prompt- and response-related evaluation workflows to test model constraints and improve answer quality through iterative refinement. Applied prompt engineering practices to explore edge cases and reduce hallucination likelihood. Collaborated on quality-focused analysis intended to strengthen model accuracy and reliability. • Conducted multi-turn constraint testing via prompt workflows • Identified edge cases and failure modes • Refined guidance to minimize hallucinations • Improved response alignment through targeted corrections

2024 - Present

Education

Work History