01Machine learning research / Berlin

ML inference,optimized at the model core.

02Where we work

From research to the runtime.

We work across the layers that determine how AI runs in production: model architecture, training, and GPU kernels.

01

Model architecture

02

Training

03

GPU kernels

03Open for conversation

We are talking to early customers and research partners.

Tell us what you are running in production.

hello@halbach.ai