Resolve AI’s SRE-1 model finds root causes of production alerts as AI agents go on call
Incident remediation company Resolve AI Inc. has announced the launch of SRE-1, a new model that has been trained to find the root cause of production alerts, and Primary On-call, a new service that puts its artificial intelligence agents on call ahead of human engineers. The new model reflects the company’s argument that while AI […] The post Resolve AI’s SRE-1 model finds root causes of…
Resolve AI has introduced SRE-1, an innovative model capable of identifying the root cause of production alerts. This powerful AI agent joins Primary On-call, a service that puts the AI ahead of human engineers, offering a proactive approach to incident remediation. Resolve AI's team of 16 AI researchers spent seven days training the 27-billion-parameter model on up to 512 GPUs, utilizing Snowflake's Cortex Training service to fine-tune the open-weight model.
Early tests show remarkable results, with SRE-1's pass rate climbing from 6.9% to 38.8% and a tenfold improvement in mean time to resolution. The AI agent can close alerts, mitigate problems where permitted, or escalate to a human when necessary, always cross-referencing the fix against the production baseline. Salesforce has already adopted Primary On-call, with their software engineering principal architect reporting that the agent successfully closed over 70% of the CRM Analytics team's paging alerts without waking any engineers.
With up to 80% of paged alerts reportedly handled without escalation, Resolve AI's solution promises significant efficiency gains for critical production incidents.
Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.