Large Language Model Intern

Brain is an early-stage startup with an ambitious vision to build exciting applications using Generative AI. The immense new power of Generative AI will make it so that everyone, not just technical programmers, can build and share exciting and interesting applications with the world. We have an incredible team, with investments led by Laurene Powell Jobs and Goodwater Capital.

As a Large Language Model Intern at Brain, you will build and/or fine-tune custom Large Language Models (LLMs) to supercharge the performance of our Imagica platform on specific application or other domains. Included in that process would be assembling and cleaning data, setting up relevant infrastructure, training and/or fine-tuning models, and evaluating performance on an appropriate data set.

How you’ll contribute to the vision:

• Help identify application or other domains where building custom LLMs would be an advantage over generic LLMs
• Collect data needed for training custom LLMs
• Clean and process the data for training/evaluations
• Train and/or find-tune custom LLMs
• Evaluate the performance of custom LLMs between each other and with respect to established generic LLMs
• Communicate findings and results to team
• Package results into a form readily usable in the Imagica platform

We’re excited about you because you:

• Are genuinely excited about the vast possibilities of Generative AI
• Want to be part of creating useful and innovative user-facing applications with Generative AI
• Experience with training and/or fine-tuning small to medium-sized (1B to 20B parameters) LLMs
• Familiar with several current open source LLMs
• Familiar with HuggingFace and other similar environments
• Experience with creating, processing, and cleaning data sets for model training/tuning/evaluation
• Experience with running evaluation metrics with held-out eval sets
• Are currently pursuing or recently completed a B.S., M.S., or Ph.D in computer science or related field
• You are fearlessly self-driven, yet enjoy helping others meet our goals.
• You enjoy solving problems and helping others meet our goals.

We know you work hard! Below are some benefits with working at Brain:

• We are a small team. You get to see many aspects of building a great product!
• We are also a growing team. We have plans to offer a lot more.
• Be a part of something massive and influence outcomes for our business.
• The ability to make, own and carry out decisions.

What to expect during our interview selection process:

• A written portion where you will discuss what you find exciting about interning at Brain and ideas you have for Imagica and how custom LLMs can help
• A face to face (remote) session where we will review your background, past projects, discuss your written exercise, and talk about how you would improve Generative AI and Imagica in particular with custom LLMs

About Brain:
Brain Technologies is based in San Mateo, California and was founded by serial entrepreneur Jerry Yue in December 2015 after founding multiple successful tech companies in Beijing including billion dollar company Benlai. Brain currently hosts its growing team in San Mateo and Chengdu, China. If you want to build great applications with Generative AI, we want to hear from you!