Inserloftdocs

Research

Research Areas

What Inserloft Research focuses on.

Inserloft Research is organized around a focused set of areas:

  • AI and LLMs broadly
  • Frontier model development, feeding directly into Kyro
  • Model architectures
  • Training methodology
  • Alignment
  • Inference efficiency
  • Agents
  • Distributed systems, which underpin both training at scale and serving models through the Platform

Inserloft is also building an internal research process inspired by how modern AI labs organize their research programs — treating research as an ongoing, structured effort rather than a series of one-off projects.