LLM Engineer (Mid-Senior)

About

We are partnering with a well-funded, rapidly scaling deep-tech company operating at the intersection of advanced AI and next-generation computing to find their next Senior LLM Engineer. Backed by strong commercial traction and global enterprise clients, the company is building highly efficient, production-grade AI systems designed to solve complex real-world problems at scale.

Their team combines world-class researchers and engineers working on cutting-edge challenges in large-scale model development, optimization, and deployment. This is an opportunity to join a highly technical environment where you will directly contribute to the future of large language models - not just apply them.

As a Senior LLM Engineer, you will design, train, and optimize large-scale transformer models, contributing across pretraining, post-training alignment (SFT, RLHF, DPO), evaluation, and inference optimization. This is a deeply technical role focused on core model development rather than downstream application or prompt engineering.

Key Responsibilities

• Design and train transformer-based models from scratch, including large-scale pretraining pipelines
• Contribute to post-training workflows such as SFT, RLHF, and DPO
• Build and optimize large-scale data pipelines for training and evaluation
• Improve model performance through architecture, training, and efficiency optimizations
• Optimize inference and training performance across GPU/HPC environments
• Collaborate with engineering teams to deploy models into production systems
• Mentor junior engineers and contribute to technical best practices

Required Experience & Skills

• 2+ years of hands-on experience training transformer or LLM models from scratch
• 5+ years overall experience for Senior-level candidates across deep learning applications
• Strong understanding of transformers, optimization, and deep learning fundamentals
• Expertise in Python, PyTorch, and the Hugging Face ecosystem
• Experience with distributed training frameworks such as DeepSpeed, FSDP, or Megatron
• Familiarity with inference optimization tools such as vLLM or TensorRT-LLM
• Experience working with large datasets, scalable training pipelines, and GPU optimization

Why Apply?

• Work on true LLM innovation, not just downstream applications
• Influence next-generation AI systems at scale
• Join a highly technical, research-driven environment with real-world impact
• Competitive compensation, flexible working, and strong growth potential

Recruiter's Note

We are specifically targeting engineers who have built and optimized models themselves—not candidates focused purely on prompt engineering or API-based LLM usage. If you have contributed to large-scale training pipelines, model optimization, or architecture-level improvements, we would like to hear from you.