Home / Companies / Northflank / Blog / Post Details
Content Deep Dive

How to run AI coding agents in AWS

Blog post from Northflank

Post Details
Company
Date Published
Author
Daniel Adeboye
Word Count
2,078
Company Posts That Month
16
Language
English
Hacker News Points
-
Post removed?
No
Summary

AI coding agents such as Claude Code, Codex, and OpenCode can run in AWS through EC2, ECS or Fargate, self-managed EKS platforms, or Northflank Cloud Harnesses deployed through a bring-your-own-cloud model. Hosting agents within an AWS account can keep code and workspaces near private infrastructure, enable access to VPC resources such as RDS and internal APIs, apply IAM permissions and CloudTrail auditing, and bill compute and Amazon Bedrock model usage through AWS. A basic EC2 deployment involves assigning a narrowly scoped IAM role, installing the required tools and agent CLI, retrieving credentials from Secrets Manager, and using tools such as tmux to preserve interactive sessions, but it requires users to manage patching, isolation, credentials, storage, team access, recovery, and idle costs. The guide presents Northflank Cloud Harnesses as an alternative that provisions isolated microVM-based agent workspaces on an EKS cluster in the customer’s AWS account or on Northflank-managed infrastructure, with persistent storage, web terminal and SSH access, configurable networking, repository integration, and support for several agent tools. For Claude Code, Harnesses can use Amazon Bedrock through configured environment variables and AWS permissions, while access to private databases and services depends on VPC routing, security groups, and appropriately scoped credentials.

Trends Found in this Post
Trend Post Mentions Total Month Mentions Posts Companies MoM
AI Coding Assistant 7 No monthly metrics for this publish month.
Secrets Management 3 No monthly metrics for this publish month.
Kubernetes 1 No monthly metrics for this publish month.
Use This Data

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.