Toloka
Human-centric AI data platform founded in 2014, backed by $72M from Bezos Expeditions, powering generative AI with crowdsourced annotation across 100+ countries.
What is Toloka?
Toloka AI is a human-centric data solutions provider founded in 2014 by Olga Megorskaya, headquartered in Amsterdam, Netherlands. Operating as a unit of Nebius Group NV, the platform combines human expertise with machine learning to deliver high-quality training data for AI systems, with a strong focus on generative AI and large language models. In May 2025, Toloka secured $72 million in funding led by Jeff Bezos' Bezos Expeditions. The platform connects a global crowd of contributors ("Tolokers") across 100+ countries with microtasks including image annotation, text classification, audio transcription, video labeling, semantic segmentation, search relevance evaluation, content quality assessment, named entity recognition, sentiment analysis, and AI model response rating. Toloka relaunched its platform in December 2024 with enhanced tools for data production, labeling, and synthetic data curation for generative AI. Contributors work flexibly from anywhere with internet access, choosing tasks at their own pace. Payments are processed via Payoneer, Skrill, and YooMoney.
Mission & values
Combine human domain expertise with machine learning to deliver training, evaluation, and safety data for frontier AI agents and large language models.
Qualifications
Toloka is accessible to beginners — no advanced background or special skills are typically required for microtasks. Key requirements include creating an account, successfully passing simple qualification tests, and accurately following task instructions. Some tasks may require specific language proficiency or be location-restricted. For specialized roles like AI Tutors or LLM annotators, professional or educational experience in data annotation, proficiency in web searching and fact-checking, and B2 or higher English are expected. Consistent, accurate work increases access to more and higher-paying tasks. Microtask earnings typically range from $0.01–$0.10 per task ($2–$8/hr equivalent), with specialized tasks offering up to $18/hr. Full-time internal roles (AI Tutor, Sales, Engineering) offer $41K–$81K annually.
Leadership
Olga Megorskaya
Co-Founder & CEO
Spent years building human-in-the-loop systems at Yandex before co-founding Toloka in 2014; led the spinout from Yandex into an independent company in 2020.
Hiring process
-
1
Sign up at toloka.ai/annotators
Create an account; verification depends on country and task pool.
About 1 day
-
2
Take onboarding training
Complete platform training and qualifying tests for specific projects.
About 2 days
-
3
Pick available tasks
Browse and accept individual microtasks (image classification, transcription, content moderation, etc.).
-
4
Complete tasks
Tasks range from a few seconds to several minutes; complete enough to clear the $20 withdrawal threshold.
-
5
Withdraw earnings
Cash out via supported methods; minimum $20 per the platform terms.
About 5 days
Funding
Company information
- Industry: AI Data & Crowdsourcing
- Location: Amsterdam, Netherlands
- Founded: 2014
- Company size: 201-1000
- Website: toloka.ai ↗
- LinkedIn: Company page ↗
- X / Twitter: Profile ↗