Open to opportunities

Griffins Ochieng

@griffinsochieng1

Message

Data annotation and LLM evaluation specialist improving model accuracy through rigorous QA.

United States

Message

What I'm looking for

I’m looking for remote, asynchronous work (15–25 hours/week) where I can apply rigorous QA, rubric-driven annotation, and LLM evaluation to improve model accuracy—especially through RLHF-style feedback and clear, actionable guidance for training teams.

I’m a detail-oriented Data Science professional with 4.5+ years of hands-on experience in data annotation, LLM evaluation, and human feedback methodologies (RLHF). I build and apply annotation rubrics, run QA audits, and deliver structured, actionable written feedback that helps teams refine model performance. I focus on factual consistency, logical reasoning, and domain-specific accuracy across multi-turn conversations.

In my current contract role, I annotate and validate 500+ text-based tasks weekly while maintaining a 98% QA score, evaluating outputs for factual accuracy, grammar, coherence, relevance, and safety. I also lead calibration efforts with 15+ remote annotators, improving team-wide IAA from 85% to 93% over 6 months, and I identify failure modes (e.g., contradictions and outdated facts) to support targeted model retraining. Previously, I labeled 10,000+ NLP datapoints and performed daily QA audits, achieving 96.5% audit accuracy, while documentation improvements reduced new-hire ramp-up time by 25%.

Experience

Work history, roles, and key accomplishments

Current

AI Data Annotator & QA Specialist

Current

Metriq AI

Jan 2024 - Present (2 years 6 months)

Annotate and validate 500+ text-based tasks weekly for an LLM training project while maintaining a 98% QA score. Evaluate outputs for factual accuracy, grammar, coherence, relevance, and safety; develop annotation rubrics and run consensus calibration to improve IAA from 85% to 93%.

LLM Evaluation Data Annotation Quality Assurance (QA)Factual Consistency Checking Response Ranking Hallucination Detection

Data Labeling Specialist

ScaleBridge Data Solutions

Jun 2022 - Dec 2023 (1 year 6 months)

Labeled and categorized 10,000+ data points across NLP tasks including sentiment analysis, entity extraction, and intent classification for chatbot interactions. Conducted daily peer annotation QA audits and used Excel macros and basic Python (Pandas) to clean and prepare datasets for labeling platforms.

Data Labeling Sentiment Analysis Data Quality Assurance Data Cleaning Excel Macros Pandas Datasets Client Reporting

Research Assistant

Cognitive Science Lab – University of California, Berkeley

Sep 2019 - May 2021 (1 year 8 months)

Conducted literature reviews across 150+ peer-reviewed journals, extracting and categorizing variables into a structured relational database. Designed experimental prompts for human-subject studies, and transcribed/annotated audio recordings in ELAN to code linguistic markers supporting research on conversational dynamics.

literature review Data Extraction Relational Databases Linguistic Marker Coding Academic Writing

Education

Degrees, certifications, and relevant coursework

University of California, Berkeley

Master of Science in Data Science, Data Science

Grade: GPA: 3.9/4.0

Activities and societies: Relevant coursework included Machine Learning, NLP, Advanced Statistics, Database Management, Data Visualization, and HCI.

Earned a Master of Science in Data Science at UC Berkeley, with a capstone thesis evaluating bias in transformer-based models using annotator demographic metadata.

The University of Texas at Austin

Bachelor of Arts in Linguistics, Linguistics

Grade: GPA: 3.8/4.0

Activities and societies: Minor in Computer Science; coursework included Syntax & Semantics, Computational Linguistics, Cognitive Psychology, and Introduction to Algorithms.

Earned a Bachelor of Arts in Linguistics at UT Austin, with honors including Dean’s List (all semesters) and the Excellence in Undergraduate Research Award.