For employers

Hire this AI Trainer

Sign in or create an account to invite AI Trainers to your job.

Invite to Job
K

Karhira Arnold

AI Operations & Quality Specialist

Italy flagRoma, Italy
$30.00/hrIntermediateInternal Proprietary ToolingOtherDon T Disclose

Key Skills

Software

Internal/Proprietary Tooling
Other
Don't disclose

Top Subject Matter

LLM evaluation
Rlhf Domain Expertise
World Knowledge

Top Data Types

VideoVideo
AudioAudio
TextText

Top Task Types

RLHFRLHF
ClassificationClassification
Text GenerationText Generation
Question AnsweringQuestion Answering
Text SummarizationText Summarization
Fine-tuningFine-tuning
Data CollectionData Collection
Prompt + Response Writing (SFT)Prompt + Response Writing (SFT)
TranscriptionTranscription
Evaluation/RatingEvaluation/Rating
Red TeamingRed Teaming

Freelancer Overview

Two years in AI evaluation at production scale, working across RLHF, LLM evaluation, video and audio assessment, preference ranking, reward modelling, and world knowledge benchmarking at Annutacon and Invisible Technologies. Operates across six evaluation divisions simultaneously on the Verity Lab platform, covering task types from dynamic effects rating and audio-visual matching to intent understanding and emotion comprehension. Took on reviewer responsibilities after consistently reaching the top of the internal quality grading system, earned through output. Has a track record of catching systemic issues others miss, including annotation errors caused by model-level defects and pattern failures in live training pipelines. Brings a STEM foundation from a BSc in Civil Engineering. Technical content, structured problem solving, and precision under pressure are built in. Remote, cross-timezone, autonomous by default.

IntermediateEnglishItalianRussianFrench

Labeling Experience

AI Operations & Quality Specialist – Annutacon / Verity Lab

VideoRLHF
As an AI Operations & Quality Specialist at Annutacon / Verity Lab, I reviewed and audited video and text model outputs across multiple evaluation divisions. I performed RLHF annotation, preference ranking, world knowledge benchmarking, and quality auditing for dynamic, audio-visual, and emotion-based model outputs. Regularly assigned reviewer duties, I ensured annotation accuracy and compliance while reporting systemic quality findings to the team. • Evaluated video model outputs for audiovisual, identity, and emotion consistency against RLHF rubrics. • Audited annotator work, provided calibration feedback, and reported systemic issues impacting data quality. • Built and reviewed benchmark questions targeting model knowledge gaps across several domains. • Ran reward modeling and preference ranking across varied content types according to project rubrics.

As an AI Operations & Quality Specialist at Annutacon / Verity Lab, I reviewed and audited video and text model outputs across multiple evaluation divisions. I performed RLHF annotation, preference ranking, world knowledge benchmarking, and quality auditing for dynamic, audio-visual, and emotion-based model outputs. Regularly assigned reviewer duties, I ensured annotation accuracy and compliance while reporting systemic quality findings to the team. • Evaluated video model outputs for audiovisual, identity, and emotion consistency against RLHF rubrics. • Audited annotator work, provided calibration feedback, and reported systemic issues impacting data quality. • Built and reviewed benchmark questions targeting model knowledge gaps across several domains. • Ran reward modeling and preference ranking across varied content types according to project rubrics.

2025 - Present

Advanced Editor, AI Trainer & Quality Specialist – Invisible Technologies

Text
As an Advanced Editor, AI Trainer & Quality Specialist at Invisible Technologies, I conducted structured annotation, prompt review, and iterative feedback on text and mixed-format tasks. I evaluated AI outputs for factuality, tone, and instruction alignment, and was selected for an internal quality task force to maintain and improve dataset integrity. My role involved workflow diagnosis and being a performance benchmark for other annotators. • Conducted structured annotation and iterative feedback on diverse dataset tasks and formats. • Reviewed AI outputs for alignment, factual correctness, and nuanced tone. • Screened high-impact datasets and flagged annotation or workflow errors. • Diagnosed and remediated real-time workflow quality issues alongside senior contributors.

As an Advanced Editor, AI Trainer & Quality Specialist at Invisible Technologies, I conducted structured annotation, prompt review, and iterative feedback on text and mixed-format tasks. I evaluated AI outputs for factuality, tone, and instruction alignment, and was selected for an internal quality task force to maintain and improve dataset integrity. My role involved workflow diagnosis and being a performance benchmark for other annotators. • Conducted structured annotation and iterative feedback on diverse dataset tasks and formats. • Reviewed AI outputs for alignment, factual correctness, and nuanced tone. • Screened high-impact datasets and flagged annotation or workflow errors. • Diagnosed and remediated real-time workflow quality issues alongside senior contributors.

2024 - 2025

Education

Z

Zaporizhia State Engineering Academy

Engineering Diploma, Engineering Studies

Engineering Diploma
2012 - 2014
C

Cyprus International University

Bachelor of Science, Civil Engineering

Bachelor of Science
Not specified

Work History

G

Grey Studios Ltd

Freelance Civil Engineer

Mbabane
2019 - 2023
S

S&B Construction

General Foreman / Junior Engineer

Matsapha
2016 - 2019