דרושים LLM Inference Engineer בכל הארץ

We’ve partnered with a very well funded (and known) company who are building the backbone of next-generation code-generation models. Not just another model-training lab, they’re re-engineering how LLMs are trained at scale: distributed systems, reliability, fault tolerance, checkpointing, GPU recovery, all at a mega scale.

Engineers here own everything end-to-end, from infrastructure to optimisation, with access to thousands of GPUs and zero bureaucracy.

It’s early enough that your technical decisions will shape the architecture and culture, but funded well enough to move fast. The team’s goal is to set the standard for code-intelligent AI systems globally. If you like solving deep technical challenges most companies can’t even simulate, this one’s worth a serious look.

Role

• Build and optimise multi-device LLM inference pipelines for source code generation.
• Work across standard transformers and custom linear attention architectures.
• Experiment with low-precision inference, tensor parallelism, and cutting-edge vendor hardware (NVIDIA, AWS).
• Contribute across the full stack — from Python/C++/CUDA/Triton to Rust-friendly rewrites and vLLM optimisations.
• Stay close to the research frontier and translate new ideas into production-grade performance gains.

Requirements

• Hands on production knowledge of LLMs.
• Understanding of transformer architectures and LLM inference.
• Strong background in distributed systems, performance engineering, or hardware acceleration.
• Hands-on experience with PyTorch or JAX, CUDA, Triton, or low-level optimisation.
• Comfort with theoretical computer science and practical systems design.
• Background in hands-on software development (non-ML)

This is a rare early opportunity to shape architecture, influence direction, and work with thousands of GPUs on problems that push the frontier of inference efficiency. You’ll have autonomy, ownership, and a direct hand in defining how AI systems scale.

Benefits

• Up to $400k + equity
• 37 days PTO
• Health insurance (including dependents)
• Learning & home office budgets
• Company equipment
• Remote-first with quarterly team trips to Paris