Principal ML Architect - GCP

Nelly Works Ltd; Greater London; Remote; GBP 120,000 - 180,000; Full time

Job summary

Nelly Works is seeking a Principal ML Architect to anchor its AI/ML practice in the UK. You’ll start hands-on, building scalable ML platforms on GCP and shaping a path to leading a dedicated ML team.

You’ll drive end-to-end architectures, engage with CTOs and client partners on pre-sales, and optimise cost and performance across Vertex AI, GKE, TPUs/GPUs, and BigQuery.

Qualifications

• Significant experience as a Principal/Senior ML Architect or Staff+ ML Engineer with production ML systems delivering measurable impact.
• Strong architecture skills across scale, security, reliability and cost.
• Deep hands-on GCP ML experience (Vertex AI, GKE, TPUs) preferred; AWS/Azure acceptable for strong architects.
• Expertise in LLM optimisation, distributed training and high-performance serving.
• Solid MLOps foundations and production agentic workflows.

Responsibilities

• Architect end-to-end ML platforms on GCP (Vertex AI, GKE, TPUs/GPUs, BigQuery).
• Optimise large-scale training and fine-tuning and high-throughput inference.
• Design production agentic systems with evaluation and guardrails.
• Build GDPR-aligned, data-sovereign AI for European customers.
• Own solution economics through capacity planning and ROI.
• Lead technical pre-sales and engage with CTOs/Heads of ML.
• Mentor engineers now and eventually hire/lead an ML team.

Skills

Principal ML Architect

GCP Vertex AI

GKE

MLOps

LLM optimisation

Pre-sales

Tools

Vertex AI

GKE

TPUs

BigQuery

Job description

Remote: UK, ideally London (no visa sponsorship)

Our client is a specialist Google Cloud consultancy founded by ex-Google Cloud engineers and leaders. They embed with customer teams to take AI from prototype to production at scale.

AI/ML is their fastest-growing practice, and they need a Principal ML Architect to anchor it. You’ll start as a hands-on senior IC. As the practice grows, you’ll have a clear path to building and leading a team of ML specialists.

What you’ll do

• Architect end-to-end ML platforms on GCP (Vertex AI, GKE, TPUs/GPUs, BigQuery), covering training, serving, MLOps and monitoring.
• Optimise large-scale training and fine-tuning (JAX, PyTorch/XLA, LoRA) and high-throughput inference (vLLM, TensorRT-LLM, quantisation).
• Design production agentic systems (LangGraph, Google ADK) with strong evaluation and guardrails.
• Build private, GDPR-aligned, data-sovereign AI for European customers.
• Own solution economics through capacity planning, cost optimisation and ROI.
• Lead technical pre-sales and act as the senior voice with CTOs and Heads of ML.
• Mentor engineers now, and in time hire and lead a dedicated ML team.

What you’ll bring

• Significant experience as a Principal/Senior ML Architect or Staff+ ML Engineer, with production ML systems that delivered measurable impact.
• Strong architecture skills across scale, security, reliability and cost.
• Deep hands-on GCP ML experience (Vertex AI, GKE, TPUs) is strongly preferred. Exceptional AWS or Azure architects will be considered.
• Expertise in LLM optimisation, distributed training and high-performance serving.
• Solid MLOps foundations and experience running agentic workflows in production.
• Clear communication with engineers and executives, and an ambition to lead a team.

Location & Travel

• Based: UK, ideally London or Greater London
• Remote: Primarily remote, with flexibility for occasional travel to meet clients, partners, and team members (conferences, offsites, collaborative sessions)

About Nelly Works

Nelly Works is aa specialist talent consultancy focusing in recruiting senior technical and commercial talent for high-growth deep tech, cloud and AI-focused organizations. We understand the nuances of building world-class teams and match exceptional talent with the right opportunities.