Orca reports 90 percent lower query costs with Bedrock pipeline

Orca Security reduced query costs by 90 percent and latency by 75 percent against its initial baseline after deploying Discovery AI on Amazon Bedrock, AWS says. The system translates security analysts' plain-language questions into queries across more than 900 cloud and application resource types. It handles thousands of queries per day with 15-second response times. The pipeline retrieves validated examples, selects only relevant resource definitions and checks generated queries against Orca's API. Structured identifiers can bypass model inference entirely. Failed validation triggers corrections, but requests that exhaust the retry limit return an invalid-filter response.