Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
Scale AI
Document
Text
Computer Code Programming
Transcription
RLHF
Computer Programming/CodingI am an experienced AI trainer and code evaluator with hands-on work across multiple LLM post-training projects at Revelo, one of the leading human data platforms for AI code training. My work spans code quality audits, behavioral data annotation, and LLM output evaluation across domains including DevOps/Infrastructure as Code, databases (MongoDB), and web development. I bring a strong technical background that allows me to assess model-generated code for correctness, reasoning, and alignment with real-world engineering standards. I have contributed to projects involving code generation evaluation, PR writing assessment, specification rubric development, and debugging analysis including both solo and review-based workflows. My ability to work across diverse technical domains and annotation task types makes me a versatile contributor to RLHF pipelines and LLM fine-tuning efforts.
This project is focused on reviewing the trajectories of two agentic models trying to solve an issue, looking for behavioral issues and then comparing both responses to determine which one was better.
The objective is to determine whether the labeler effectively guided the AI through a multi-turn conversation to complete a task, while providing high-quality comparative evaluations between two competing model responses at each interaction. For code review, the goal is to create and evaluate realistic interactions (prompts) focusing on asking for code review and repository structure analysis.
In this project you'll interact* 10+ times with a coding agent that will write two different solutions. Your job is to create realistic prompts and evaluate the responses focused on planning the architecture of a complex system.
Bachelor of Science, Computer Science
Senior Fullstack Engineer
Full Stack Engineer
