AI safety testing in Telugu and English: provoke and document jailbreaks, bias and harmful output from chat models, and judge the quality of their Telugu answers. Evaluation skill matters more than security experience. Remote hourly contract at $16–22/hour, weekly pay; H-1B and STEM OPT excluded.
Odia and English AI safety work: probe chat models for jailbreaks, bias and harmful output and judge whether their Odia answers hold up. One of the lowest-resource languages in the family and the busiest Indian variant, with 88 hired this month. Remote hourly contract, $16–22/hour.
Review AI-narrated audiobooks in European Portuguese and flag where the synthetic narrator goes wrong: dropped or added words, bad pronunciation, misread numbers and abbreviations, unnatural rhythm. For native speakers from Portugal who listen to audiobooks. Remote hourly contract, around 20 hours a week, $15–20/hour.
AI safety work in Kannada and English: test chat models for jailbreaks, bias and harmful output, and judge whether their Kannada answers are accurate and appropriate. Unlike the European variants, prior red-teaming is not listed as a requirement. Remote hourly contract, $16–22/hour, weekly pay.
Audit Brazilian Portuguese speech data for Amazon's Sonic project: check transcripts against audio with fixed error codes and pass/fail verdicts, and fix word-level timestamp alignment. Native Portuguese as spoken in Brazil is a hard requirement; rationales are written in English. Remote hourly contract at $25/hour.
Quality-check AI-narrated audiobooks in German (Germany): pinpoint mispronounced, missing or extra words, wrongly read numbers and abbreviations, unnatural prosody and audio artefacts, and judge the overall listen. For native German audiobook listeners. Remote hourly contract, about 20 hours a week, $15–20/hour.
Native Odiya (Oriya) speakers in India annotate the layout of real Odiya PDF pages and transcribe each text region exactly in Odiya script, handwriting included, to build training data for document AI. Remote hourly contract at $12.68/hour, with every task reviewed by a second Odiya expert.
Native Malayalam speakers in India segment real Malayalam PDF pages into typed, ordered regions and transcribe every one exactly in Malayalam script, handwriting included, for document AI training data. Remote hourly contract at $12.68/hour; each task gets a full second-expert review.
Evaluate AI-narrated audiobooks in Dutch (Netherlands): mark each mispronounced, skipped or added word, misread number, awkward intonation and audio glitch, then rate the overall listen. For native Dutch speakers who are regular audiobook listeners. Remote hourly contract, about 20 hours a week, $15–20/hour.
Listen to AI-narrated audiobooks in Brazilian Portuguese and tag where the synthetic narrator slips: mispronunciations, skipped or extra words, wrong readings of numbers and abbreviations, flat or odd intonation. For native Brazilian audiobook listeners. Remote hourly contract, around 20 hours a week, $15–20/hour.
Test AI models in Gujarati and English for jailbreaks, bias and harmful answers, and judge whether their Gujarati is accurate and appropriate. Evaluation judgment is the core requirement. Open to the Gujarati diaspora, with no residence rule published. Remote hourly contract at $16–22/hour; 56 hired this month.
Listen to AI-narrated Italian audiobooks and log every slip in the synthetic voice: skipped or mispronounced words, misread numbers, odd intonation, glitches. For native Italian speakers who actually listen to audiobooks. Remote hourly contract, about 20 hours a week, $15–20/hour.
Assess AI-narrated audiobooks in French (France) and annotate each failure: wrong or missing liaisons, mispronounced or skipped words, misread numbers and abbreviations, stilted intonation, glitches. For native French speakers who love audiobooks. Remote hourly contract, around 20 hours a week, $15–20/hour.
Review AI-narrated audiobooks in US / North American English and flag where the synthetic narrator slips: mispronounced names, dropped or added words, numbers and abbreviations read wrongly, flat intonation, glitches. The best-paid version of this project at $20–25/hour. Remote hourly contract, about 20 hours a week.
Listen to AI-narrated audiobooks in US / North American Spanish and log each narration error: mispronounced, skipped or added words, misread numbers and abbreviations, stiff intonation, audio glitches. For native Spanish-speaking audiobook listeners. Remote hourly contract, about 20 hours a week, $15–20/hour.
Audit Hindi speech data for Amazon's Sonic collections: check annotators' transcriptions against the audio with fixed error codes and pass/fail verdicts, and correct word-level timestamp alignment. Native Hindi (India) is a hard requirement; rulebook and rationales are in English. Remote hourly contract at $18/hour.
Probe AI models in Tamil and English for jailbreaks, bias and harmful output, and assess whether their Tamil answers are accurate and appropriate. Evaluation judgment is the core requirement, not prior red-teaming. Remote hourly contract, $16–22/hour, weekly pay; 57 hired this month.
Audit Mandarin Chinese speech data for two Amazon Sonic collections: judge annotators' transcripts against audio using fixed error codes, and fix word-level timestamp alignment. Native Mandarin as spoken in mainland China is a hard requirement; all rules and rationales are in English. Remote hourly contract at $21.50/hour.
Audit Korean speech data for Amazon's Sonic project: check transcripts against audio with fixed error codes and pass/fail verdicts, and correct word-level timestamp alignment. Native Korean as spoken in South Korea is a hard requirement; rationales are written in English. Remote hourly contract at $25/hour.
Audit Japanese speech data for Amazon's Sonic collections: verify annotators' transcripts against the audio with fixed error codes and pass/fail calls, and correct word-level timestamp alignment. Native Japanese as spoken in Japan is a hard requirement; rationales are in English. Remote hourly contract at $37.50/hour.
Test AI chat models in Malayalam and English for safety failures and judge whether their answers are accurate, complete and appropriate. Native Malayalam plus careful evaluation skills; prior red-teaming is a plus rather than the core requirement. Remote hourly contract at $16–22/hour, paid weekly.
Native Telugu speakers in India map the layout of real Telugu PDF pages (headings, tables, figures, reading order) and transcribe every text region exactly in Telugu script, including handwriting, to train document AI. Remote hourly contract at $12.68/hour, with peer review of every task.
Audit French speech data for Amazon's Sonic project: check annotators' transcripts against audio with fixed error codes and pass/fail calls, and correct word-level timestamp alignment. Native French as spoken in France is a hard requirement; rationales are in English. Joint top rate of the Sonic set at $41.50/hour, remote contract.
Native Gujarati speakers in India annotate the structure of real Gujarati PDF pages and transcribe every text region exactly in Gujarati script, handwriting included, to build training data for document AI. Remote hourly contract at $12.68/hour; every task is peer-reviewed by a second Gujarati expert.
New roles land most days. Get them on Telegram or Discord as they are added, or read how the platforms pay before you apply.