Senior ML Engineer

Foundation Model Leadership | Applied AI Research & Systems

• Location:
  
  London, UK (on-site preferred; remote with monthly visit possible)
• Job Type: Full-time | Senior
• Compensation: £100,000 – £200,000 + share options
• Our Client: VC-backed AI/ML start-up

• Must have the right to work in the UK or be able to obtain it
• On-site presence in London preferred
• Remote candidates must commit to at least monthly London visits
• Visa sponsorship not available

Our client is developing a novel foundation model to enable fully automated, unsupervised software delivery in embedded control systems. As a VC-backed, early-stage AI company based in West London, they are building the core ML stack from first principles. This is a hands-on technical leadership role focused on architecting, optimizing, and deploying large-scale foundation models in a high-urgency, high-impact environment.

• Lead research, development, and production deployment of the foundation model
• Define long-term technical strategy for high-performance ML systems
• Optimize models across diverse hardware environments
• Architect scalable distributed training and inference pipelines
• Build GPU-accelerated components, including custom CUDA kernels
• Profile and optimize the full ML stack end-to-end
• Create internal tooling, benchmarks, and evaluation harnesses
• Work closely with founders to translate product goals into technical roadmaps

• Python and CUDA C/C++
• PyTorch (preferred) or similar deep learning frameworks
• Distributed training and large-scale inference systems
• Cloud platforms (AWS, Azure, or GCP)
• Modern foundation model architectures (MoE, state-space models)

• Proven experience designing and implementing large-scale foundation models
• Strong hands-on performance optimization and debugging skills
• Practical experience with distributed training and inference
• Deep knowledge of modern deep learning frameworks
• Experience operating ML systems in production environments
• High urgency, ownership mindset, and comfort with ambiguity

• Custom CUDA kernel development
• Internal ML tooling or benchmarking experience
• Containerisation and orchestration exposure
• Embedded or control systems background
• Experience at top-tier AI/ML companies or research labs

• Ground-floor role shaping a novel foundation model
• Direct collaboration with founders
• High technical ownership and strategic influence
• Streamlined interview process (2 stages)
• Transparent, no-jargon engineering culture