Browse categories
Explore
Fiverr Pro
English
$
USD
I will build scalable ai and llm infrastructure on AWS using terraform
AWS Cloud Architect and Infrastructure Specialist
Hi, I'm Nisal, an AWS & Azure Cloud Architect. I design secure, highly available cloud infrastructure, automate it with Terraform, and build private AI/LLM (RAG) environments. With smart FinOps, I cut...
Level 2
Has met high performance criteria and has a proven track record for meeting client expectations.
About this Gig
Ready to scale your Generative AI models to production?
Deploying Large Language Models (LLMs) requires secure and cost-effective cloud architecture. I specialize in building production-grade AI and LLM infrastructure on AWS entirely through Terraform. Whether you are deploying open-source models (Llama 3, Mistral) or building enterprise MLOps pipelines with AWS SageMaker and Bedrock, I will automate your cloud environment.
What I Deliver:
- Cloud Networking: Custom VPCs, private subnets, and strict IAM roles.
- LLM Deployment: Provisioning Amazon EC2 GPU instances, ECS, or EKS clusters.
- Managed AI: Infrastructure for SageMaker endpoints and AWS Bedrock.
- APIs: Application Load Balancers and API Gateways for global access.
- Monitoring: CloudWatch integration for GPU tracking and cost control.
Why Work With Me?
- IaC Expert: Clean, modular, and well-documented Terraform source code.
- Cost-Optimized: Architecture designed to prevent wasted GPU spend.
- DevOps Focus: Deep expertise in AWS, Docker, GitHub CI/CD, and MongoDB data layers.
️ Please send me a message before ordering to discuss your model requirements!
Tools:
Docker
•
GitHub
•
BitBucket
•
Cloud Formation
•
Hashicorp Vault
Framework:
Terraform
Cloud Provider:
Amazon Web Services
Programming language:
Python
Expertise:
Installation
•
Development
•
Configuration
My Portfolio
FAQ
Which Large Language Models (LLMs) can you help deploy?
I can deploy almost any Generative AI model you need. This includes open-source models like Llama 3, Mistral, Falcon, and HuggingFace models hosted on AWS EC2 GPU instances or Amazon EKS. I also set up managed infrastructure for AWS Bedrock and Amazon SageMaker endpoints.
Do I get to keep the Terraform source code?
Yes! Every package includes the complete, modular Terraform source code (.tf files). This means your infrastructure is 100% yours, version-controlled, and easy for your team to replicate or modify in the future.
Will my AI infrastructure be secure?
Absolutely. I strictly follow the AWS Well-Architected Framework for MLOps. Your models will be deployed in custom VPCs with private subnets, strict IAM roles, and security groups ensuring your data and inference endpoints are never exposed to the public internet without proper API gateways.
Do I need to provide my AWS login details?
No, you should never share your root password! I will guide you on how to create a temporary IAM user with specific, restricted permissions (or provide an Access Key / Secret Key pair) so I can securely run the Terraform deployment on your account.
Can you help me scale if my Generative AI app goes viral?
Yes! If you select the Standard or Premium packages, I configure Application Load Balancers and Auto-Scaling Groups (or EKS clusters). This ensures that as your traffic increases, AWS will automatically spin up new GPU instances to handle the load, and scale them down when not in use to save you mon
What if I am not sure which AWS services I need for my model?
No problem at all. Just send me a message before placing an order! Tell me what model you want to run and your expected user traffic, and I will recommend the most cost-effective AWS architecture (EC2 vs. SageMaker vs. Bedrock) for your specific use case.
