Core Systems Engineer (Distributed LLM Engine & Mesh)

Let’s save each other some time. This is not an "AI wrapper" project.

We are not looking for API integrators, prompt engineers, or data scientists who simply import transformers. If your expertise is built on top of someone else's black-box APIs, you will hate this role.

We are engineering a novel LLM inference/training engine from scratch, designed to run over a distributed mesh network. We need foundational builders who understand the architectural difference between an LLM model (the weights) and an LLM engine (the execution environment), and know exactly how to optimize the latter.

You should apply if your mental models are built on:

• First-Principles: Math, physics, and information theory.
• Systems Architecture: You understand tensor operations, memory bottlenecks, and how matrix multiplication actually maps to the hardware.
• Distributed Topologies: You can design for P2P mesh networks, parallel processing, and asynchronous consensus across high-latency, unreliable nodes.

The Reality:

• Location: 100% Remote, anywhere on the planet.
• Compensation: This is an equity-only, ground-floor position. We are building foundational infrastructure.

If you want to build the engine that everyone else will eventually run their models on, let's talk.

Izvor: https://rs.linkedin.com/jobs/view/core-systems-engineer-distributed-llm-engine-mesh-at-cognition-commons-4440491250

Nivo: mid level