You’re a machine learning engineer who is an expert at productionizing and optimizing models.
We have a huge library of community-contributed machine learning models. You’ll maintain some of the most popular ones so they’re fast and reliable.
It’ll involve packaging open-source models with Cog, optimizing them, giving them user-friendly APIs, making them trainable, and monitoring them over time.
We’re looking for the right person, not just someone who checks boxes, so you don’t need to satisfy all of these things. But, you might have some of these qualities:
- A balance of software engineering and machine learning skills.
- You can squeeze every last drop of performance out of a GPU.
- You can turn a machine learning paper into working code.
- You’ve worked with model compression techniques like pruning and distillation.
- You know your way out of
CUDA error: device-side assert triggered
.
- Ideally you’re involved in the generative AI community and familiar with diffusion models and similar techniques.
- You don’t need a PhD or know how to build new architectures from scratch.
- Excellent communication skills. We think most of being a programmer is not programming. We want you to be able to communicate complex topics clearly, write down your thinking, write good docs, etc.
Email us: jobs@replicate.com