Veronicah Wairimu

Key Skills

Software

Sama

Other

Top Subject Matter

Multimodal AI Data Quality

Speech & NLP AI Training

LLM & NLP Model Evaluation

Top Data Types

Image

Audio

Text

Document

Top Task Types

Classification

Transcription

Freelancer Overview

Data Quality Analyst. Core strengths include Sama and Other. Education includes Advanced Diploma, Serein Learning Institute (2023) and Certificate, Pivotech Institute of Professional Studies (2014). AI-training focus includes data types such as Image, Audio, and Text and labeling workflows including Classification, Transcription, and Evaluation.

Labeling Experience

Data Quality Analyst

SamaImageClassification

Ensured the quality and accuracy of annotated datasets across multiple data types to support large-scale AI models. Performed systematic reviews of labeled images, text, audio, and video data, maintaining data integrity. Collaborated with clients and technical teams to resolve data issues and improve annotation processes. Produced high-quality, unbiased datasets for robust AI model training. • Conducted quality checks and delivered actionable insights. • Achieved consistently high data quality scores above 98%. • Applied root cause analysis to recurring issues and implemented corrective actions. • Ensured compliance with security and governance standards.

Ensured the quality and accuracy of annotated datasets across multiple data types to support large-scale AI models. Performed systematic reviews of labeled images, text, audio, and video data, maintaining data integrity. Collaborated with clients and technical teams to resolve data issues and improve annotation processes. Produced high-quality, unbiased datasets for robust AI model training. • Conducted quality checks and delivered actionable insights. • Achieved consistently high data quality scores above 98%. • Applied root cause analysis to recurring issues and implemented corrective actions. • Ensured compliance with security and governance standards.

2010 - Present

Audio Transcriptionist

OtherAudioTranscription

Transcribed Swahili and English audio files with high fidelity for AI training initiatives. Managed large volumes of audio data under stringent quality and formatting requirements. Produced clear, accurate transcripts used for AI and NLP tasks to enhance model understanding. Consistently met tight deadlines while ensuring transcript quality for natural language processing. • Followed strict quality and formatting guidelines. • Produced transcripts for use in AI dataset construction. • Supported high-volume transcription under fast turnaround times. • Facilitated effective Swahili-English language model training.

Transcribed Swahili and English audio files with high fidelity for AI training initiatives. Managed large volumes of audio data under stringent quality and formatting requirements. Produced clear, accurate transcripts used for AI and NLP tasks to enhance model understanding. Consistently met tight deadlines while ensuring transcript quality for natural language processing. • Followed strict quality and formatting guidelines. • Produced transcripts for use in AI dataset construction. • Supported high-volume transcription under fast turnaround times. • Facilitated effective Swahili-English language model training.

2026 - 2026

Evaluation Specialist

OtherText

Evaluated large language model (LLM) outputs to identify errors, bias, hallucinations, and safety concerns. Employed structured evaluation frameworks and taxonomy-based classification to create specialized training datasets. Supported reinforcement learning and model fine-tuning through curated evaluations. Collaborated remotely with global teams and adapted to evolving AI training guidelines. • Applied evaluation and classification frameworks to LLM outputs. • Generated datasets for reinforcement learning and model optimization. • Flagged edge cases and safety risks to improve model robustness. • Maintained consistency with dynamic project requirements.

Evaluated large language model (LLM) outputs to identify errors, bias, hallucinations, and safety concerns. Employed structured evaluation frameworks and taxonomy-based classification to create specialized training datasets. Supported reinforcement learning and model fine-tuning through curated evaluations. Collaborated remotely with global teams and adapted to evolving AI training guidelines. • Applied evaluation and classification frameworks to LLM outputs. • Generated datasets for reinforcement learning and model optimization. • Flagged edge cases and safety risks to improve model robustness. • Maintained consistency with dynamic project requirements.

2025 - 2026

Education

Work History