Frequently Asked Questions

Loevech is a professional AI data annotation and quality rating platform. We connect human evaluators (annotators) with businesses and research teams that need high-quality labeled data to train, fine-tune, and evaluate AI models. Our platform supports text, image, audio, and video content across multiple rating methodologies.

Anyone can sign up to become an annotator on Loevech. No prior experience in AI or data annotation is required -- every project comes with clear guidelines and instructions. Domain expertise (e.g., in law, medicine, linguistics, or technology) is valued for specialized projects and may unlock higher-paying opportunities.

Click "Get Started" on the homepage and choose whether you want to join as an annotator (to rate data and earn) or as a business (to post annotation projects). You can sign up with your email or Google account. The process takes less than 2 minutes.

Signing up and browsing projects is completely free for annotators. Businesses can create projects with flexible pricing based on the volume and complexity of their annotation needs. There are no minimum project size requirements.

Tasks vary by project and include: text sentiment analysis, image classification, audio quality evaluation, video content review, A/B comparison of AI outputs, multi-label categorization, Likert scale quality ratings, and binary approve/reject decisions. New projects are added regularly across diverse content types and domains.

Completely flexible. There are no minimum hours, fixed schedules, or deadlines for individual annotators. You can work whenever you want, from anywhere with an internet connection. Simply log in, browse available projects, and start rating at your own pace.

Your personal analytics dashboard shows your total tasks completed, projects contributed to, average time per task, and quality scores. Quality is measured through inter-rater reliability (agreement with other annotators) and performance on gold-standard tasks. Strong performance unlocks access to more projects.

No special equipment is needed. All you need is a device with a web browser (computer, tablet, or phone) and a stable internet connection. For audio and video projects, headphones are recommended but not required.

Yes! You can browse and contribute to as many active projects as you want. Each project has its own guidelines, progress tracking, and task queue, so you can easily switch between projects based on your interests and expertise.

You can create projects for text, image, audio, and video content using four rating methods: Likert scales (configurable 3-10 points), binary approve/reject, A/B side-by-side comparison, and multi-label classification with custom tags. Each project supports custom guidelines, configurable labels, and quality benchmarks.

Quality is built into every level of the platform: (1) Inter-rater reliability scoring measures agreement between annotators. (2) Gold-standard tasks with known answers track individual annotator accuracy. (3) Real-time analytics dashboards let you monitor quality metrics as results come in. (4) Required ratings per task ensure sufficient consensus.

Tasks can be added in three ways: (1) Manual entry through the project interface. (2) CSV upload for bulk imports. (3) File upload for image, audio, and video content. Each task can include the content to rate, optional comparison content (for A/B tasks), and reference titles.

Yes. All project results can be exported as CSV files with complete data including individual ratings, aggregated scores (averages, majority votes, label counts), rater information, and time-spent metrics. This data is ready for direct use in your AI training pipeline.

No. Unlike many competitors, Loevech has no minimum project size requirement. You can create a project with as few or as many tasks as needed. This makes the platform ideal for both large-scale production annotation and small research pilots.

Loevech offers several key advantages: (1) No minimum project size -- start with any volume. (2) Built-in quality control with inter-rater reliability scoring. (3) Real-time performance analytics for every annotator and project. (4) Support for text, image, audio, and video across multiple rating methods. (5) Simple, intuitive interface with zero setup complexity. (6) Fair, transparent compensation for annotators.

All data is encrypted in transit and at rest. Our platform runs on enterprise-grade cloud infrastructure with SSL/TLS encryption, secure authentication (including two-factor), and role-based access controls. We follow GDPR compliance practices and never share your data with third parties.

Inter-rater reliability (IRR) measures the degree to which different annotators agree on their ratings for the same content. High IRR indicates consistent, reliable annotations. Loevech calculates and displays IRR metrics in real-time so both annotators and project managers can monitor and improve data quality.