ML Systems Engineer

Overview

You join a fast-growing AI infrastructure startup to own software-side modelling and benchmarking that demonstrates system viability. Working with the CTO, you will build a software ‘digital twin’ to evaluate behaviour ahead of hardware, and develop workloads such as agentic AI and GraphRAG to show measurable improvements. You will maintain a rigorous benchmark suite and design experiments to quantify the impact of the semantic memory layer on inference. Your work supports pilots, fundraising, and technical validation, with close collaboration with hardware teams.

Responsibilities

• Own the software model (digital twin) for system evaluation
• Build agentic AI and GraphRAG workloads to demonstrate improvements
• Develop and maintain a benchmark suite (latency, GPU utilisation, token reduction, throughput, cost per query)
• Design experiments isolating semantic memory layer impact on inference performance
• Develop enterprise knowledge graph datasets and evaluation methodologies
• Coordinate with hardware/systems teams to align software models with hardware capabilities
• Generate evidence to support pilots, fundraising, and technical validation

Key requirements

• Commercial experience in AI systems, retrieval, or AI infrastructure with production software experience
• Hands-on experience with agentic pipelines, LLM fine-tuning, RAG/GraphRAG, or knowledge graphs
• Strong Python proficiency; comfortable across ML, distributed systems, and performance engineering
• Track record building benchmarks/eval frameworks with rigorous methodology
• Systems thinker with high agency; comfortable with ambiguity
• Strong communicator who can translate technical results into clear evidence

• systems thinker
• high agency
• comfortable with ambiguity
• Python
• AI systems / retrieval / AI infrastructure
• agentic pipelines

…