Senior ML Systems Engineer (Inference) – Hybrid + Tokens

TensorX; Dublin; Hybrid; EUR 120,000 - 180,000; Full time

Benefits offered by this job

Hybrid working from Dublin office

25 days paid annual leave

Free inference tokens

Job summary

Tensorix in Dublin is seeking a Senior ML Systems Engineer (Inference) to own the model serving layer, deploying and tuning open-source LLMs on on-prem GPU fleets and AWS. This hands-on role shapes performance, cost and reliability of token delivery.

You will work with vLLM, SGLang and TensorRT-LLM on NVIDIA hardware, drive hardware planning, benchmarking, and collaborate with platform and product teams. Hybrid work from Dublin with remote flexibility, 25 days leave, and a competitive package.

Qualifications

• 5+ years of professional experience in ML infrastructure or production ML workloads.
• Hands-on experience deploying and tuning large language models with modern inference frameworks.
• Strong knowledge of GPU architecture and NVIDIA hardware (H100/H200).
• Experience with inference optimisation techniques and benchmarking methodologies.
• Proficiency in Python and systems-level code reading/writing.
• Familiarity with Linux, containers and orchestration of GPU workloads.

Responsibilities

• Deploy and operate open-source LLMs in production using vLLM, SGLang, TensorRT-LLM and similar frameworks.
• Profile and tune inference workloads for latency, throughput and GPU utilisation.
• Benchmark new models and maintain internal tooling for reproducible results.
• Lead GPU procurement planning and capacity planning.
• Build and maintain GPU infra spanning on-prem and cloud workloads.
• Instrument serving stack with metrics and drive reliability improvements.
• Prototype new optimisation techniques and stay current with research.

Skills

ML infrastructure

System engineering

Python

Linux

GPU architecture

CUDA fundamentals

Benchmarking

Communication

Ambiguity tolerance

Education

BSc/MSc in CS or related

Tools

vLLM

SGLang

TensorRT-LLM

Kubernetes

AWS

Job description

Tensorix in Dublin is seeking a Senior ML Systems Engineer (Inference) to own the model serving layer, deploying and tuning open-source LLMs on on-prem GPU fleets and AWS. This hands-on role shapes performance, cost and reliability of token delivery.

You will work with vLLM, SGLang and TensorRT-LLM on NVIDIA hardware, drive hardware planning, benchmarking, and collaborate with platform and product teams. Hybrid work from Dublin with remote flexibility, 25 days leave, and a competitive package.