Machine Learning Intern

We refuse to be ordinary, so neither is this job description! At the core of our work is our belief that teachers hold the power to move educational outcomes forward. We want technical experts who not only care deeply about solving complex problems, but also about impact. It is our responsibility to include the people impacted during model design, consider human engagement risks, and protect our students. So if you’re down to push on how ML models can improve equity in education, let’s talk! 

‍

What we’re looking for and why we need you!

We are seeking an eager Machine Learning Intern in our pursuit of bringing computer vision-language-enabled solutions to education. Too many AI solutions in education require students to sit on Chromebooks using a janky UX to interact with an AI chatbot. Not for our kids, thank you – to us, that’s a dystopia. We envision a world where students challenge each other through discussion, try hard at hard problems, and build core knowledge. For AI to have a role in the world we envision, the AI needs to be able to diagnose student work using optical sensors, freeing students from the constraints of the keyboard. 

There’s no more natural gauge of whether you understand a math problem than working it out with nothing more than a question and a pencil. We’re working on a 3-year work study funded by the Bill & Melinda Gates Foundation to collect a large sample of handwritten student work, hire math content experts to diagnose the work, and train an AI to replicate the expert diagnosis. We are looking to expand the capacity of our team of engineers to build models capable of making the nuanced distinction between correct and partially correct work that can handle image inputs that combine numbers, figures, and text. It’s a hard problem filled with exciting opportunities to push the state of the art. The goal is to build delightful products and experiences for millions of students, teachers, and families.

We’re looking for Machine Learning Interns to join us as we build products that will give teachers the benefit of an AI co-pilot without requiring students to sit at Chromebooks. This is a 12 week internship, with a tentative start date of November 1, 2023. 

‍

At EdLight, You Will:

• Curate datasets to use in fine tuning multi-modal LLMs
• Experiment with adding multi-modal capabilities to open source LLMs to better understand where the challenges will be in training data at scale
• Laugh and play as part of a team that is tackling hard problems with humility

‍

You Are:

• 3+ years of college-level education in pursuit of a degree in Computer Science or an applicable field
• Seeking experience that will make a positive difference in the world where you can apply your skills, learn and grow
• Passionate about educational equity and creating solutions that make quality instruction accessible for all
• Proficient in Python and object-oriented programming (OOP) 
• Captivated by the applications of machine learning to handwritten student work
• Proficient in state-of-the-art natural language processing (NLP) and Computer Vision (CV)
• Constantly searching for ways to remain up to date with state-of-the-art machine learning research

‍

The perks

• $25 - $50 / hour (commensurate with education and experience)
• Minimum of 10-15 hours a week
• Flexible, remote working environment
• Culture: We are a small, yet agile and mighty team. If you are searching for a close-knit team of coworkers driven by educational equity and tech accessibility in schools, we hope that you reach out!

‍

If you are interested in this role, please complete our application form (linked here).