Graduate AI Trainer @
Hatch - Sydney, Australia
Apply NowJob Description
This is a Graduate AI Trainer (Frontier AI Lab) role with Hatch based in Sydney, NSW, Australia== Hatch ==Role Seniority - GraduateRole: Graduate AI Trainer (Frontier AI Lab)Location: RemotePay range: $60 per hourType: Flexible / Project-BasedFocus: University graduates in various fields (Including but not limited to business, engineering, communications, law)The OpportunityHatch is partnering with a world-leading AI Lab to build the next generation of AI. We aren''t looking for top graduates from professional degrees to help train AI models on general tasks.You will be part of an elite team of AI trainers ready to be deployed into high-impact projects where you will teach AI models how to think follow complex instructions, and solve real-world problems.What You''ll Be DoingAs an AI Trainer, you will act as a human-in-the-loop training how AI models and agents think.While the specific tasks vary by project, your core responsibility is to apply your first-principle thinking and personal judgment to AI outputs.Instruction Following: Execute complex, multi-step tasks within digital environments exactly as briefed.Precision Evaluation: Review AI-generated responses for accuracy, professional tone, and logical consistency.Data Refinement: Identify subtle errors that programmatic tools miss, nuances in language, professional etiquette, or logical flow.Feedback Integration: Adapt quickly to new guidelines as the AI Lab iterates on their models.Who You AreWe are looking for recent university graduates who are detail oriented, motivated to do their best work, proactive and excellent communicators.This Role Is For You IfYou Have "The Eye": Your attention to detail is non-negotiable. You''re the person who catches the missing comma, the broken link, or the slight shift in brand voice that everyone else misses.You Are a strong first-principles thinker: You know how to break down problems and solve them rationally.You Are Unquestionably Reliable: We have a zero-tolerance policy for fraud and have an incredibly high bar for your the way you show up for a client project.You are intuitive: You don''t just click buttons. You understand the intent behind an instruction and ensure the output matches that intent.Core RequirementsEducation: An undergraduate degreeCommunication: Exceptional written English and the ability to explain why a piece of data is incorrect.Tech Stack: Comfortable working across multiple browser tabs, specialized interfaces, and communication tools (Slack/Discord).Availability: Ability to join a "ready-pool" for upcoming pilots (various shift patterns available).Why become a Hatch AI trainerShape the Frontier: Work on the actual models that will define the next decade of technology.Flexibility: Once cleared for the team, you''ll be first in line for project-based work that fits your schedule.Competitive Edge: Gain experience in "AI Alignment," one of the fastest-growing sectors in the global economy.Note on Quality Control: > Our partner labs prioritize human-led quality control. Your work will be frequently "spot-checked" by senior reviewers. High-quality contributors will be prioritised for long-term, high-volume engagements.Before we jump into the responsibilities of the role. No matter what you come in knowing, you''ll be learning new things all the time and the Hatch team will be there to support your growth.Key Responsibilities Executing simulated tasks Reviewing AI outputs Identifying errorsKey Strengths Attention to detail First-principles thinking Exceptional communication Tech-savviness Adaptability Team collaborationA Final Note: This is a role with a Frontier AI Lab that is working with Hatch.
Created: 2026-02-18