Revelo
Revelo works exclusively on training and evaluation data for code. The engineers, tools, and review process used for client datasets also produced the tasks in the Code Index.
Focused on code
Our work covers code training data, executable evaluations, and analysis of agent runs. The infrastructure and review process are built for software tasks, including repository setup, test design, trajectory review, and technical quality control.
Experienced engineers
Tasks are written and reviewed by vetted senior developers with current production experience. Revelo matches projects with engineers by language, framework, and domain, then manages the work from a small pilot through larger releases.
The same network wrote and reviewed the tasks in this index. Read the methodology →
From sourcing to delivery
Revelo manages the operational work around a dataset or evaluation:
- Project management
- Tooling & data handling
- Quality management
- Expert sourcing & vetting
- Compliance & payment
- Technical contributor operations
Talk to us
Revelo works with model labs on:
- Training datasets for code generation
- Evaluations of new and unreleased models
- Leaderboard analysis, including failure modes and agent trajectories