The provider's government and enterprise clients get their data jobs done at any scale, with updates shipped in minutes.
The provider, a data management partner to federal agencies and large companies, wanted its applications to scale with its clients' work, stay secure and receive updates quickly.
What changed
What stood in the way.
- A monolith that ran its work through cron jobs could not scale, and tracking security events across many AWS services was hard.
- Its own database proxy machines were costly to keep highly available, and slow build containers delayed every deployment.
- 30-40%lower costs (as reported)
- 50%faster resolution of security events (as reported)
- Up to 66%shorter database failover (as reported)
- 40%less deployment time (as reported)
What we built.
- A queue-based architecture with containers and Redis queues, so workers scale out for each kind of job.
- A central security posture tool with AWS Config, Prowler, GuardDuty and Inspector watching for threats.
- Amazon RDS Proxy in place of the self-managed database proxy.
- Caching in CodeBuild containers and automated, standard deployments with AWS CodeDeploy.
- AWS containers
- Redis queues
- AWS Config
- GuardDuty
- Amazon RDS Proxy
- AWS CodeDeploy
What this work shows about us.
The fix was structural: work moved from scheduled jobs to queues with workers that scale, rather than bigger servers.
Every security event lands in one place, so the team resolves them in half the time (as reported).
Want to move from cron jobs to a queue-based architecture on AWS? Tell us what you want to change. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with the right team.
Start your plan ↗If we built it today, with AI.
This work came before today's AI. If the US grants management and software provider asked us now, this is what we would add, step by step.
An agent would read each new security finding, group duplicates, explain the risk in plain words and draft the fix for an engineer. Engineers decide and apply every change.
An agent would check a failed build or deployment, find the cause in the logs and propose the fix. People still approve every release.
Where this work fits.
Facts and figures from the project records; nothing added.
Want this for your business?
Tell us what you run or what you're building. Our AI assistant, Ask Unthinkable, helps you define your requirements, then connects you with a senior AI engineer.
Send us a note instead
An AI agent. This prototype uses prepared answers and sends nothing.
We use essential cookies to run this site. With your consent we also use analytics cookies to see which pages help. Privacy policy