Key Skills

Software

Scale AI

Top Subject Matter

No subject matter listed

Top Data Types

Image

Text

Top Task Types

Classification

Evaluation Rating

Object Detection

Prompt Response Writing SFT

RLHF

Freelancer Overview

I'm an experienced AI trainer and data evaluator with nearly 1,000 completed tasks at Outlier.ai, specializing in reinforcement learning from human feedback (RLHF), prompt evaluation, and multimodal vision annotation. My work involves reviewing AI-generated content for quality, coherence, and safety while following strict annotation protocols under tight deadlines. I've completed over 30 certification modules covering technical research from companies like Anthropic, OpenAI, and Google, building deep expertise in AI evaluation frameworks and prompting techniques. My experience spans multiple project types including Cypher Evals (prompt evaluation and response quality rating), Cypher RLHF (reinforcement learning with human feedback), and Vision SFT (image-caption pair alignment and labeling). I've also served as a reviewer for quality control, validating other contributors' work. Beyond annotation, I'm actively applying my AI knowledge to real-world product development through independent SaaS projects. I'm bilingual in English and French, with strong copywriting skills and daily experience using advanced AI platforms like ChatGPT, Claude, Gemini, and Perplexity. My proven track record includes earning nearly $8,000 through high-volume, high-accuracy task completion, demonstrating both my efficiency and precision in AI training data work. What sets me apart is my combination of technical AI evaluation expertise, practical application experience, and ability to maintain consist

Labeling Experience

Vision SFT

Scale AIImagePrompt Response Writing SFT

This project involves writing prompts that refer to pictures, then creating strong responses that incorporate details from those images to improve AI's ability to process and understand visual content. (each tasks can take between 45 min. to 3h to complete)

This project involves writing prompts that refer to pictures, then creating strong responses that incorporate details from those images to improve AI's ability to process and understand visual content. (each tasks can take between 45 min. to 3h to complete)

2025 - 2025

Cypher Evals

Scale AITextEvaluation Rating

This project consists of evaluating based on multiple criterias and ranking the best of the two LLM-generated responses. (each tasks can take between 15 min. to 45min to complete)

This project consists of evaluating based on multiple criterias and ranking the best of the two LLM-generated responses. (each tasks can take between 15 min. to 45min to complete)

2025 - 2025

Cypher RLHF

Scale AITextRLHF

This project consists of writing natural, non-contrived prompts with specific use cases per task (e.g., extraction, classification, creative writing, open QA, etc.), sometimes accompanied by reference text, then evaluating and ranking the responses generated by the model. (each tasks can take between 15 min. to 1h to complete)

This project consists of writing natural, non-contrived prompts with specific use cases per task (e.g., extraction, classification, creative writing, open QA, etc.), sometimes accompanied by reference text, then evaluating and ranking the responses generated by the model. (each tasks can take between 15 min. to 1h to complete)

2024 - 2025

Education

Work History