How we cut our cloud costs by 99%

Kate Whitcomb
April 10, 2020
facebook twitter In

TLDR: We didn’t move our computing to the edge! We built a groundbreaking, developer-first cloud platform to help more efficiently stream and analyze massive amounts of data in real-time. Learn more here.

Longer story: We had no choice. As a startup selling thousands of our cutting-edge smart baby monitors in 2018-2019, we were faced with cloud compute costs reaching $70 per customer per month. Adding computer vision to a baby monitor sounded great - until we got the bill! At the rate we were going, we were about to spend millions per month on Amazon Web Services products, particularly Kinesis and Lambda/Fargate. When we explored alternatives to help us manage our streaming data analytics costs with similar performance, there were none.

So, we built one, and brought our costs down to <$1 per customer per month.

Traditionally, doing real time analytics on streaming data in the cloud has been perceived as complex and prohibitively expensive. This has limited the use of real-time analytics in the cloud for many businesses and presented a major white space in the cloud computing market.

Amazon’s prices are high for a reason- this is a complex problem. We had no choice but to solve it.

Introducing Chrysalis Cloud, the first serverless end-to-end data streaming and analytics cloud platform. How can you save money with Chrysalis Cloud?

Our solution is designed for companies with many data streams and a desire to perform real-time analysis on that data. This is particularly useful for customers managing continuous video or audio streams and continuous sensor data from an IoT device.

For example, our first customer is a security camera company using cloud-based AI for people tracking and risk management. They have been constrained by the costs of the current streaming and hosting products.

How does it work?

Our service will be offered in two parts: the Chrysalis Data Streaming service and the Chrysalis Data Processing service. We can offer both as a managed service or as an a la carte solution, all in a developer-first, quick-to-deploy stack.

The Chrysalis Cloud includes the following proprietary innovations:

A developer-first platform to easily stream any data (including video) from any streaming device to the cloud and process data in real-time. A custom-built container orchestrator with substantially lower management overhead (compared to Google’s Kubernetes). A method of utilizing the edge vs. cloud to allow for more accurate data collection and reduced streaming. A system that dynamically assigns compute resources per user depending on the level of service that the user is subscribing to. A proprietary method of managing compute to reduce variability in operating costs, further adding to the efficiency of our stream processing system. A custom application and infrastructure monitoring service that ensures high availability. We’d love to work with you!

The new Chrysalis Cloud platform is compute-first, cloud agnostic, scalable, and has a low operating footprint compared to existing services. This new infrastructure is capable of real-time data streaming, SSL encryption for all streams, storage, real-time compute capability, real-time logging and more.

Our offering to consumers will help you save up to 70% compared to compute solutions offered by AWS or Google Cloud Platform. This opens the door for companies to perform more advanced functions, including real time AI, with far greater efficiency and without the need for expensive servers or specialty edge devices.

If you are interested in learning more, please visit our website,, and sign up. We will be happy to tell you more!

Kate Whitcomb

Hi everyone! Exciting news. We've launched a cloud computing platform! Save 70% on your AWS bills. Keep reading!