Kévin Berger

Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Computer Code Programming

Text

Top Task Types

Audio Recording

Computer Programming Coding

Prompt Response Writing SFT

RLHF

Translation Localization

Freelancer Overview

As a French native AI trainer and senior reviewer, fluent in English and Spanish, I bring hands-on experience across multiple high-impact AI training projects. I specialize in finance, STEM (math and coding), and voice AI, combining technical precision with linguistic fluency. My work includes generating and reviewing prompts, correcting AI-generated outputs, annotating complex data, and refining transcriptions and dubbing scripts for naturalness and clarity. I’ve been promoted to Senior Reviewer on major projects at Outlier (Scale AI) due to the consistent quality of my work and my ability to follow detailed guidelines while mentoring other contributors. With a background in financial consulting and university-level teaching in Excel/VBA, I combine analytical rigor with a strong attention to detail, ideal for delivering high-quality data annotation at scale.

Labeling Experience

Generalist LLM Evaluation – Senior Reviewer

Scale AITextRLHFEvaluation Rating

As part of a large-scale project aimed at training and fine-tuning general-purpose language models, I was promoted to Senior Reviewer based on the consistent quality of my evaluations. My responsibilities included: - Reviewing AI-generated responses across a wide range of topics (general knowledge, logical reasoning, creative writing, etc.). - Assessing accuracy, coherence, tone, and helpfulness in alignment with evolving guidelines. - Comparing model outputs (side-by-side evaluation), ranking completions, and identifying failure modes. - Reviewing the work of other annotators and providing quality control and mentoring support. - Liaising with project admins to ensure guideline updates were well implemented.

As part of a large-scale project aimed at training and fine-tuning general-purpose language models, I was promoted to Senior Reviewer based on the consistent quality of my evaluations. My responsibilities included: - Reviewing AI-generated responses across a wide range of topics (general knowledge, logical reasoning, creative writing, etc.). - Assessing accuracy, coherence, tone, and helpfulness in alignment with evolving guidelines. - Comparing model outputs (side-by-side evaluation), ranking completions, and identifying failure modes. - Reviewing the work of other annotators and providing quality control and mentoring support. - Liaising with project admins to ensure guideline updates were well implemented.

2024

LLM Trainer – SOTA Coding Prompts & Model Evaluation

Scale AIComputer Code ProgrammingRLHFFine Tuning

In this STEM AI training project, here are my main missions: - Designing high-quality prompts in coding, aiming to challenge and evaluate LLM reasoning. - Evaluating AI outputs, identifying reasoning errors, inaccuracies, or hallucinations in code solutions. - Scoring and correcting model responses for clarity, correctness, and pedagogical value. - Generating model-like responses (SOTA – State of the Art) that serve as ideal answers to complex technical questions. - Writing reviewer notes to contributors to explain mistakes and guide future improvements in model alignment.

In this STEM AI training project, here are my main missions: - Designing high-quality prompts in coding, aiming to challenge and evaluate LLM reasoning. - Evaluating AI outputs, identifying reasoning errors, inaccuracies, or hallucinations in code solutions. - Scoring and correcting model responses for clarity, correctness, and pedagogical value. - Generating model-like responses (SOTA – State of the Art) that serve as ideal answers to complex technical questions. - Writing reviewer notes to contributors to explain mistakes and guide future improvements in model alignment.

2024

AI Voice Acting & Recording

Scale AIAudioClassificationText Generation

In this project, here are my main missions : - Recording high-quality audio using professional equipment, in a quiet environment with no background noise, reading prompts in a natural and expressive tone, following specific guidelines (intonation, pacing, emotional delivery, etc.). - Ensuring clarity and accuracy of pronunciation, articulation, and consistency across takes. - Adapting the voice style based on the intended use case (e.g., conversational, informative, instructional). - Providing or reviewing transcripts, labeling speech segments, or correcting misalignments between audio and text (speech-to-text alignment tasks). - Following detailed technical specs, including naming conventions, file formats, and platform workflows. - Reviewing contributors recording at a Senior level.

In this project, here are my main missions : - Recording high-quality audio using professional equipment, in a quiet environment with no background noise, reading prompts in a natural and expressive tone, following specific guidelines (intonation, pacing, emotional delivery, etc.). - Ensuring clarity and accuracy of pronunciation, articulation, and consistency across takes. - Adapting the voice style based on the intended use case (e.g., conversational, informative, instructional). - Providing or reviewing transcripts, labeling speech segments, or correcting misalignments between audio and text (speech-to-text alignment tasks). - Following detailed technical specs, including naming conventions, file formats, and platform workflows. - Reviewing contributors recording at a Senior level.

2024

Education

Work History