Verita AI

Verita AI is an AI infrastructure company founded in 2025, with roots at Stanford and offices in San Francisco. The company builds the "data layer" for multimodal intelligence, specializing in curating high-fidelity image, audio, video, and interaction datasets evaluated by top creators, designers, and domain experts. Verita AI's mission is to integrate human taste, safety, and rigor into next-generation AI systems, ensuring models reflect human insight, judgment, and authenticity rather than just mirroring facts. Services include Reinforcement Learning from Human Feedback (RLHF), Supervised Fine-Tuning, custom environment development, and user feedback loops to align models with human preferences. The company partners with recording studios and artists for premium music and voice data, and builds quality pipelines with expert evaluation, red teaming, and bias checks. All roles are fully remote and contract-based with flexible scheduling.

Qualifications vary by role. For Data Annotators, candidates must be US-based, at least 18 years old, with reliable internet, strong attention to detail, and the ability to follow annotation guidelines. For Audio Annotation roles, expertise in sound design or related fields and strong listening skills are required. For Remote Language Speaking Contributors, native-level fluency in the target language is essential, along with a quiet recording environment and reliable internet for recording conversational audio clips. For Machine Learning Engineer roles, proficiency in Python and TensorFlow is expected, with experience at the intersection of research and applied ML. Android QA Tester / Data Annotator roles require completing tasks on a personal Android phone. The company emphasizes working with creative professionals, practitioners, and domain experts who can bring human judgment and taste to AI training data. No formal degree is universally required, but domain expertise and practical skills are valued.