LLM Infrastructure Engineer
Hire Hangar · Remote · Argentina - Buenos Aires · Parugauy - Asuncion · Ecuador - Guayaquil · Bolivia - Santa Cruz de la Sierra · Egypt - Cairo · South Africa - Johannesburg · Columbia - Bogotá · Columbia - Medellín · Chile - Santiago · Uruguay - Montevideo · Peru - Li ·
COMPENSATION
$30k – $36k
Listed on the employer's job board (Ashby)
- Fully remote (US hours)
- LLM Infrastructure
- Python + AWS
- GPU optimization
- Senior level
This position focuses on designing and scaling the infrastructure behind large language models. You will collaborate with machine learning teams to ensure models run reliably and cost-effectively in production environments.
What you'll do
- Design infrastructure for training, fine-tuning, and serving LLMs
- Deploy inference services focusing on latency and throughput
- Manage GPU resources and optimize workload scheduling
- Build data pipelines supporting model training and evaluation
- Implement monitoring and alerting for system health
- Optimize infrastructure costs while maintaining performance
- Automate deployments using infrastructure-as-code and CI/CD
- Collaborate with researchers to productionize models
- Identify bottlenecks and resolve infrastructure issues
What you'll need
- Proven experience building LLM infrastructure is mandatory
- Hands-on experience deploying ML models in production
- Strong Python skills with backend or systems development background
- Experience with Docker and Kubernetes orchestration
- Solid understanding of distributed systems and GPU computing
- Fluent written and spoken English communication
- Prior remote work experience with US or UK companies required
Nice to have
- Strong AWS experience including EC2, EKS, S3, and SageMaker
- Background in enterprise-scale engineering environments
- Familiarity with vLLM, TensorRT-LLM, or Triton Inference Server
- Experience with Terraform or CloudFormation
- Knowledge of PyTorch Distributed, DeepSpeed, or Ray
Stack
- AWS
- Python
- Docker
- Kubernetes
- Terraform
- CloudFormation
- PyTorch
- vLLM
- TensorRT-LLM
- Triton Inference Server
How you'll work
- Fully remote position
- Must work within US time zones (EST to PST)
- Requires proficiency with remote collaboration tools
Hiring process
- Complete the full application form
- Record a video as the first interview step
- Receive email confirmation if successful