Generative AI – Generalist

Surge



<aside> 🧠 Our mission at Surge is to build the human infrastructure to power the next generation of AI.

</aside>

Surge was founded by former ML engineers to expand the frontiers of AI. We’re building a platform that powers leading AI groups, including OpenAI, Anthropic, Meta, and Google, by providing high-quality human feedback data to evaluate and train their models. Our product has been a “game-changer” for ML teams, and we’ve run a profitable business from day one without raising venture funding.

You’ll join a small, rapidly growing team of engineering leaders and ML practitioners from Google, Meta, Twitter and Airbnb. In this role, you will help make leading models, like Gemini, Llama2, and Claude, better. You’ll work closely with senior researchers and leaders, at some of the biggest companies in the world, to help shape the future of AI models.

We’ve found the best candidates have a wide range of seniority — entry level to FAANG Directors — and come from any field — tech, hard sciences, finance, and the arts and music — some without any experience in AI. The characteristics we’ve found most important: being a compelling and interesting writer and communicator, possessing intense self-motivation, and demonstrating an exacting attention to detail.

Things you’ll do:

  • Work closely with leading AI researchers to find ways to improve their models
  • Design and execute data collection projects, including developing interfaces, establishing quality control processes, managing throughput, and delivering data on time
  • Communicate with senior executives at leading AI companies and within our team
  • Codify data collection systems to make our processes more efficient and scalable

About you:

  • Exceptional communicator
  • Tenacious and scrappy
  • Demonstrated leader

To apply, please email careers@surgehq.ai with a resume and 2-3 sentences describing your interest in Surge. We love personal projects and writings too!

To apply for this job please visit surgeai.notion.site.