Expert AI Data Trainer (Mathematics)

North America - Remote /
The Agency: Worldwide Work-Sharing – Operations - Enterprise Operations /
Part-Time
What is Invisible

Website : http://www.inv.tech/
Blog: https://inv.tech/blog
Overview / Sales Deck - https://invtech.docsend.com/view/6kp3ixp
Recorded Demo - https://www.youtube.com/playlist?list=PL4135pGQh8yvHRwKatYrMDuKL7ODHBXHI
Core Values: https://www.notion.so/invisibletech/d8add47d17584715a62aa4d46f0270b8?v=187ce27ec02f4bfaa2aed9f3f2c6ad6d 

Who We Are
Invisible Technologies helps fast-growing companies scale efficiently, so they can improve margins and capture market share at the same time. We partner with our clients to run their business operations using our modern outsourcing platform to deliver efficiency, quality, and flexibility at scale.

This role will be empowering AI trainers to train AI models to solve complex problems in order to provide value to humanity. We believe that Expert AI trainers are the key to unlocking the potential of our models to be as accurate and helpful as experts. You will a part of building a system that uses experts’ guidance to teach our models how to understand difficult questions and execute complex instructions. This will foster collaboration between AI and humans that aligns with our shared vision and values.

This position will be either a 10hr/week or 20hr/week commitment depending on your availability.

We Believe That
Mission. Invisible is a world-changing company and the Operations Team is responsible for creating the systems that will launch Invisible into the stratosphere.
Ownership. Every partner on the team is the “CEO” of their area - each of us are empowered  to jump in and solve the problems in front of us.
Communication. Consistent feedback is key - we are addicted to learning and getting better.
Transparency. What one of us knows, all of us should know.
Self-Improvement. Every new mistake is a learning opportunity.
Meritocracy. The value someone produces should be reflected back in the position of responsibility and compensation that they receive.

About the Role
You will train AI towards general intelligence. Our ability to train increasingly capable and safe models depends on our ability to collect high-quality data, as well as combining data and compute with human supervision to align the models towards desired behaviors. The process of aligning AI requires supervision from many humans, including experts.

At present we are looking for domain expertise within the mathematics field.

In this role, you will:

    • Generate questions related to your domain of expertise that require expertise in the domain in order to answer correctly
    • Ask the models those questions and refine your questions
    • Evaluate the models’ answers in order to help them improve
    • Develop a framework for asking questions to ensure appropriate breadth and depth
    • Communicate your findings to researchers, engineers, and operations teams
    • Provide at least 10 hours but no more than 20 hours per week

You might thrive in this role if you have:

    • Deep domain expertise in a given field
    • Undergraduate-level mathematical education
    • Excellent written and verbal communication skills
    • The ability to formulate a structured approach to ambiguous problems
    • A curious mind and a kind personality

Job Type: Contract, 10-20 hours per week dependent on your availability

Required language: English
Compensation & Benefits
* Working Times: US (EST or PST) Hours
* $25 - $30 per hour
* Flexible Work Schedule
* Work Remotely