Founding ML Engineer — RL & Scale Training

Axiōma Search; Greater London; Hybrid; GBP 70,000 - 110,000; Full time

Job summary

Axiōma Search is seeking a skilled ML engineer to advance post-training methods for large-scale AI systems. You will improve training loops from rollout to policy updates, enhance throughput and GPU efficiency, and diagnose bottlenecks across distributed environments.

You’ll bridge ML research with systems engineering to ensure reliable, scalable model training. You will work on reinforcement learning and related post-training strategies, using RL approaches like PPO, GRPO and SDPO, and you’ll

Qualifications

• Strong programming and quantitative problem-solving skills.
• Hands-on experience with PyTorch and model training.
• Understanding of reinforcement learning or LLM post-training.
• Experience with distributed training and GPU systems.
• Strong experimental judgement.
• Ability to work comfortably across ML research and systems engineering.

Responsibilities

• Build and improve supervised fine-tuning, preference optimisation and RL methods.
• Work with approaches including PPO, GRPO and SDPO.
• Own training loops from rollout generation through to policy updates and checkpointing.
• Improve training throughput, GPU utilisation and memory efficiency.
• Profile and fix bottlenecks across distributed training.
• Investigate instability and differences between training and inference.

Skills

Strong problem-solving

Programming ability

Experimental judgement

ML research & systems

Tools

PyTorch

Distributed training

GPU systems

Job description

Axiōma Search is seeking a skilled ML engineer to advance post-training methods for large-scale AI systems. You will improve training loops from rollout to policy updates, enhance throughput and GPU efficiency, and diagnose bottlenecks across distributed environments.

You’ll bridge ML research with systems engineering to ensure reliable, scalable model training. You will work on reinforcement learning and related post-training strategies, using RL approaches like PPO, GRPO and SDPO, and you’ll