LLM Ops Engineer — Deploy, Scale & Observe AI Models

Innotech; Viseu; On-site; EUR 55,000 - 75,000; Full time

Job summary

InnoTech, a leading IT company in Portugal, seeks a Mid-Senior LLM Engineer to operationalize and deploy large language models across industries such as banking, insurance, and retail.

You will drive end-to-end model lifecycle—from installation and fine-tuning to deployment and monitoring—while collaborating with software engineering and cloud teams to ensure secure, scalable pipelines.

Qualifications

• Minimum of 2 years' experience in similar roles.
• Strong programming skills in Python, Bash, and Go.
• Hands-on experience with LLM frameworks like Hugging Face Transformers, PEFT, DeepSpeed, Megatron-LM, vLLM, or TensorRT-LLM.
• Expertise in containerization, orchestration, and cloud infrastructure using Docker, Kubernetes, Helm, Terraform, or Ansible.
• Knowledge of distributed training frameworks such as PyTorch, Ray, Horovod, or Accelerate.
• Fluency in Portuguese and residence in Portugal.
• CPLP nationality or an EU work permit is required to facilitate employment.

Responsibilities

• Operationalize large language models (LLMs) from installation and fine-tuning to deployment and monitoring.
• Ensure scalable, secure, and efficient deployment of LLMs using enterprise data.
• Collaborate at the intersection of MLOps, software engineering, and cloud infrastructure.
• Maintain governance, observability, and performance standards for production LLMs.
• Develop and optimize pipelines for training, deployment, and model management.

Skills

Python

Bash

Go

Education

BSc or MSc in Computer Science/Engineering/IT

Tools

Docker

Kubernetes

Helm

Terraform

Ansible

Hugging Face Transformers

PEFT

DeepSpeed

Megatron-LM

vLLM

TensorRT-LLM

PyTorch

Ray

Horovod

Accelerate

Job description

InnoTech, a leading IT company in Portugal, seeks a Mid-Senior LLM Engineer to operationalize and deploy large language models across industries such as banking, insurance, and retail.

You will drive end-to-end model lifecycle—from installation and fine-tuning to deployment and monitoring—while collaborating with software engineering and cloud teams to ensure secure, scalable pipelines.