We are looking for driven individuals who excel at operating computers to evaluate our MACROHARD products. You’ll get to work with the world’s latest AI Agents observing and impacting their behavior. This position is ideal for attentive individuals with a knack for thinking outside the box in software QA-style testing of AI agents in a fast-paced, innovative environment.
You’ll receive exclusive early access to our top internal models and the opportunity to test their robustness across tasks ranging from spreadsheets and slide editors to video games.
Responsibilities
- Delivering high-quality data and annotations for scenarios involving MACROHARD, and testing Computer Use Agents in digital environments.
- Identifying subtle bugs, failure modes, and unexpected agent behaviors during testing sessions to help improve Computer Use models.
- Assisting in designing and improving annotation tools tailored for MACROHARD data, agent evaluation, and QA workflows.
Required Qualifications
- Exceptional Computer Literacy
- Comfort with recording audio, video, screen sessions, or interaction logs for detailed data collection and analysis.
- Proficiency in reading and writing informal and professional English.
Preferred Qualifications
- Passion to work on the frontier of AI Agents.
- Some technical background in software, automation, tools, or related areas
- Strong communication, interpersonal, analytical, and organizational skills.
- Exceptional reading comprehension, meticulous attention to detail, and the ability to exercise autonomous judgment with limited data.
- Clever problem-solving mindset with a talent for thinking outside the box, especially in identifying and testing non-obvious scenarios, edge cases, and potential failure points.
- Passion for technological advancements, innovation in AI agent testing, rigorous QA practices, and high-quality data labeling.