I returned to AWS and was reminded why I left
AIThis post was created with the assistance of artificial intelligence (AI).

TL;DR

A long-time AWS advocate details their reasons for leaving the platform due to frustrations with complexity, cost, and policies. They recently returned briefly for testing, highlighting ongoing issues but also the platform’s capabilities.

A former AWS advocate, who left the platform after over a decade of dissatisfaction, has recently logged back in for testing purposes, citing ongoing issues with cost, complexity, and policies.

The individual, an early supporter of AWS, described their initial enthusiasm for the platform’s revolutionary cloud computing capabilities, which allowed startups to run their own systems without data centers. Over time, their relationship with AWS deteriorated due to numerous frustrations, including the lack of native client libraries for six years, high costs associated with DynamoDB, complex billing practices, and an overly complicated IAM system.

They also criticized AWS’s handling of open-source projects, accusing the company of pushing proprietary alternatives like OpenSearch and DocumentDB to capture market share, despite community opposition. The user highlighted the vendor lock-in created by AWS Lambda and expressed regret over their previous commitment, citing the difficulty of moving away from these services.

Recently, the user returned to AWS briefly to conduct research and benchmarking tests, including using AWS Bedrock for AI model testing and launching a high-core EC2 instance. They confirmed that the platform still functions as needed for specific tasks, but maintained that their overall opinion remains critical due to persistent issues and high costs.

Why It Matters

This story underscores the ongoing challenges faced by long-term AWS users, including cost management, platform complexity, and policy decisions. It highlights that even dedicated advocates can become disillusioned, which may influence enterprise and individual users considering cloud providers. The user’s return for testing also suggests that AWS’s capabilities still attract users for specific technical needs despite broader dissatisfaction.

Amazon

AWS EC2 high core instance

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background

Since its inception, AWS has revolutionized cloud computing, enabling startups and enterprises to deploy infrastructure rapidly. However, over the years, criticism has grown regarding its high costs, vendor lock-in, and complex ecosystem. Many early supporters have expressed frustration over AWS’s evolving policies and technical limitations, leading some to leave the platform entirely. This recent re-engagement reflects a nuanced view: while disillusioned, some users still see value for particular tasks.

“I left AWS after over 15 years because of the complexity, cost issues, and policy decisions. But I returned briefly for testing and benchmarking.”

— former AWS advocate

“AWS is ridiculously complex and expensive, and their policies often feel like they’re designed to lock users in and maximize profit at the expense of flexibility.”

— the user

“Even with all the issues, AWS still offers capabilities that are hard to match, which is why I needed to test it again for specific research purposes.”

— the user

Amazon

AWS Bedrock AI model testing

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What Remains Unclear

It remains unclear whether the user’s brief return indicates a potential shift in their overall stance towards AWS or if it was solely for specific testing needs. Additionally, it is uncertain how representative this experience is of broader user sentiment, as opinions about AWS vary widely among different segments.

Amazon

cloud computing benchmarking tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

What’s Next

The user plans to conclude their testing soon and will decide whether to continue using AWS for future projects. Observers will watch if their experience influences their overall view or prompts further engagement with the platform. AWS’s ongoing policy and cost management strategies may also impact future user perceptions.

Amazon

AWS cost management tools

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Why did the user leave AWS after so many years?

The user left due to frustrations with AWS’s high costs, complex billing, vendor lock-in, and technical limitations like lack of native client libraries and overly complicated IAM systems.

What prompted the user to return to AWS?

They returned briefly to conduct specific research, testing AI models on AWS Bedrock, and benchmarking high-performance hardware, which are tasks best suited to AWS’s infrastructure.

Does this mean the user is now a supporter of AWS again?

Not necessarily. The user’s return was limited to specific testing activities and does not indicate a full reversal of their previous criticisms or disillusionment with the platform.

Are their criticisms common among AWS users?

Many long-term users have expressed similar frustrations regarding cost, complexity, and vendor lock-in, though experiences vary widely depending on use cases and organizational size.

You May Also Like

The Core Ultra 7 270K was too good, so Intel scrapped the flagship Core Ultra 9 290K Plus — benchmarks of the 290K prototype find slim 2% faster performance in gaming and applications

Intel has scrapped the flagship Core Ultra 9 290K Plus, citing disappointing performance in leaked benchmarks, favoring the more balanced Core Ultra 7 270K Plus.

Linux Eliminates the Strncpy API After Six Years of Work, 360 Patches

Linux 7.2 officially eliminates the strncpy API after six years and 362 patches, replacing it with safer alternatives for kernel development.

The Real Cost of a Local-Inference Rig in 2026

Analyzing the hardware costs and considerations for running large language models locally in 2026, including VRAM constraints and value strategies.

South Korea floats using AI profits for ‘people’s dividend’

South Korea plans to use AI-driven semiconductor profits to fund a ‘people’s dividend,’ a redistributive measure aimed at easing economic inequality.