CrowdGen

Help advance AI development by contributing personal photo data! We’re collecting 20 old photos from individuals across various age groups for use in training AI systems. Submit a total of 20 personal photos: 10 photos taken within the past 2 years and 10 photos taken more than 2 years ago. Each photo must include the approximate date it was taken. Task volume will be assigned by the project manager depending on availability.

Help advance AI development by contributing personal photo data! Individuals across various age groups will submit 20 old photos for use in training AI systems. Participants will submit 20 personal photos: 10 photos taken within the past 2 years and 10 photos taken more than 2 years ago. Each photo must include the approximate date it was taken. If any submission doesn't meet requirements, it will be returned for revision.

Evaluate the relevance between search queries and the user’s intent to refine ad targeting and brand visibility in e-commerce platforms. Insights will directly impact the accuracy of product recommendations and search results. Must be a native Korean speaker with strong English proficiency and hold a Bachelor’s degree or higher. Experience in search relevance scoring and familiarity with e-commerce platforms are preferred. Able to work at least 4 consecutive hours on weekdays, responsive, reliable, and able to follow a consistent schedule.

Seeking experienced multilingual professionals for a short-term, remote project evaluating translation quality. As a quality evaluator, you'll review translated text, assign quality scores based on fluency and clarity, and tag potential errors. This role requires native Persian and Arabic proficiency, strong reading and writing skills, and the ability to work 3–4 hours daily. You must provide a valid passport, language proficiency certificate, and resume with translation experience.

Be offered consistent, daily work as part of a long-standing project with Appen. Contribute to enhancing the user experience for various platforms such as Apps, Games, Streaming TV/Movies, Music, and Podcasts. Enjoy competitive pay rates, attractive bonuses, and a diverse range of tasks that keep the work engaging. With Project Wells, you'll have the flexibility to choose your tasks and work schedule.

We're looking for native Russian speakers to take part in a two-phase language data project that supports the advancement of speech recognition technologies. Phase 1 involves evaluating and labeling Russian audio recordings based on guidelines. Phase 2 involves transcribing and correcting audio. The role requires native proficiency in Russian, comfort with audio review and transcription, and a commitment of 3 hours per day, while working remotely with flexible hours.

Evaluate the accuracy and quality of machine-translated content. Your work will support the creation of high-quality multilingual datasets by assessing translations and scoring them based on semantic similarity. Your task is to evaluate how well a translated paragraph preserves the meaning of its original version. This involves checking the semantic alignment between the source and translated texts, ensuring that the core ideas, information, and intent remain intact after translation.

Evaluate how well a translated paragraph preserves the meaning of its original version by checking the semantic alignment between the source and translated texts, ensuring that the core ideas, information, and intent remain intact after translation. This involves reading and comparing both the source and translated paragraphs on a sentence by sentence basis, focusing strictly on semantic meaning (not grammar, fluency, or tone).

We are looking for native Japanese speakers with strong English proficiency to join Project Presales, a high-impact initiative focused on improving English audio transcription quality. This is a remote, part-time role ideal for bilingual professionals with a linguistic background. Bring your language expertise to a meaningful project aimed at enhancing global speech recognition systems.

As a Social Media Evaluator, you will assess AI-generated annotations across multilingual tasks by evaluating model responses in two formats: Head2Head comparison – rating two outputs side by side and Standalone evaluation – assessing individual responses for quality. In this role, you will analyze inputs that may include images, textual queries, data, and AI-generated responses to determine accuracy, alignment, and overall user satisfaction.

Be part of a global AI initiative and help shape the future of voice technology. We're seeking native French speakers from Belgium to participate in an exciting, paid voice data collection and evaluation project. Your contributions will directly support the development of smarter, more inclusive AI systems used worldwide.

Review pairs of inputs and AI-generated outputs, then assess the quality of those outputs based on accuracy, relevance, and clarity—following the project’s established guidelines. Your evaluations will directly support improvements to model performance and help shape the future of responsible AI development.