AI Teammates: how monday.com runs production AI agents on Amazon Bedrock
Software company monday.com operates production AI agents called AI Teammates using Amazon Bedrock inside a decade-old code base. Internal company metrics show that nine in ten Builders now use AI coding tools monthly, up from roughly half a year ago. These deployment efforts have resulted in per-engineer PR throughput increasing by more than half.
Key Takeaways
- A publication from AWS Machine Learning outlines how monday.com runs production AI agents named AI Teammates on Amazon Bedrock.
According to internal production metrics from monday.com, developer usage has expanded rapidly, with nine in ten Builders adopting AI coding tools each month compared to roughly half a year ago.
- This widespread adoption has driven significant performance improvements, raising per-engineer PR throughput by more than half.
To achieve these results, the engineering team implemented architectural retrofits designed to integrate agentic systems into a decade-old code base.
- The system also relies on a confidence-scored merge play to advance toward full autonomy.
This production deployment demonstrates how engineering organizations can adapt legacy infrastructure to support scaled AI agents.
- Nine in ten Builders at monday.com utilize AI coding tools every month, showing adoption growth from roughly half a year ago.
Deploying AI agents on Amazon Bedrock has helped increase per-engineer PR throughput by more than half.
- The deployment utilizes a confidence-scored merge play to help close the gap toward full autonomy.
A publication from AWS Machine Learning outlines how monday.com runs production AI agents named AI Teammates on Amazon Bedrock. According to internal production metrics from monday.com, developer usage has expanded rapidly, with nine in ten Builders adopting AI coding tools each month compared to roughly half a year ago. This widespread adoption has driven significant performance improvements, raising per-engineer PR throughput by more than half.
To achieve these results, the engineering team implemented architectural retrofits designed to integrate agentic systems into a decade-old code base. The system also relies on a confidence-scored merge play to advance toward full autonomy. This production deployment demonstrates how engineering organizations can adapt legacy infrastructure to support scaled AI agents.
Nine in ten Builders at monday.com utilize AI coding tools every month, showing adoption growth from roughly half a year ago. Deploying AI agents on Amazon Bedrock has helped increase per-engineer PR throughput by more than half. Integrating AI Teammates into monday.com required specific architectural retrofits to support a decade-old code base.
For more details please read the original article at AWS Machine Learning.
Continue Learning
Comments
Comments appear only after moderation. Your email identifies your submission to the moderator and is never displayed here.
No approved comments yet.