LLM Engineer (SFT / RLHF / Post-Training)

About the role

We are seeking an LLM Engineer to join a leading global tech platform focused on post‑training, reasoning, and alignment of large language models. The role offers the chance to work on cutting‑edge AI systems at massive scale, primarily in a remote setting.

Key responsibilities

• Develop and maintain LLM post‑training pipelines (SFT, DPO, RLHF).
• Improve model reasoning, alignment, and reliability.
• Train and optimise models ranging from 7B to 100B+ parameters.
• Build agent systems, tool‑use, and multi‑step reasoning workflows.
• Operate large‑scale GPU clusters and implement distributed training strategies.

Required profile

• Hands‑on experience with LLM fine‑tuning techniques such as SFT and LoRA.
• Familiarity with RLHF, DPO, PPO, or GRPO methods (any is a plus).
• Experience with models like LLaMA, Qwen, Mistral, or similar.
• Strong background in LLM, NLP, or applied machine‑learning systems.

Required skills

• Python
• PyTorch
• DeepSpeed
• LoRA
• SFT
• RLHF
• DPO
• PPO
• GRPO
• LLaMA
• Qwen
• Mistral
• GPU clusters
• Distributed training

What we offer

• Remote work arrangement.
• Opportunity to work on state‑of‑the‑art LLM systems at global scale.
• Fast‑moving, high‑impact environment.
• Exposure to next‑generation AI research in reasoning, agents, and alignment.