Founding Platform Engineer

The Role

Our client is hiring a Founding Platform Engineer to build the backbone of their training, evaluation, and inference systems

This is a central role within the business. You’ll be responsible for designing and building systems that transform messy, large-scale data into formats that models can reliably learn from. You’ll operate across the full stack from data ingestion through to production inference.

What You’ll Work On

• Building and scaling infrastructure for model training, evaluation, and inference
• Designing high-performance inference systems
• Creating clean, standardised dataset interfaces
• Developing ingestion and processing pipelines across multiple data sources
• Building indexing, search, and retrieval systems for large-scale datasets
• Ensuring datasets are versioned, reproducible, and reliable
• Improving performance across distributed systems (GPU, networking, storage)
• Identifying and resolving bottlenecks in ML pipelines
• Implementing fault detection and recovery mechanisms
• Building internal tooling for debugging, inspection, and data quality
• Collaborating closely with ML engineers and researchers
• Contributing to best practices around handling sensitive data

What We Are Looking For:

• Backend, infrastructure, or distributed systems engineering
• Scaling data pipelines or production systems
• Large-scale ML systems (training and/or inference)
• GPU performance, memory constraints, and data loading challenges
• Designing systems for datasets that exceed memory limits
• Indexing, search, or retrieval systems
• Balancing latency, throughput, and cost trade-offs
• Building internal tools for technical teams
• Working with sensitive or regulated data
• A strong focus on reliability, reproducibility, and simplicity
• Modern infrastructure (cloud, containers, Kubernetes, CI/CD, observability)
• Writing clean, maintainable code with pragmatic decision-making
• Thriving in fast-paced, early-stage environments