Resolve AI, a production-operations startup supported by Greylock and Lightspeed Venture Partners, has announced a major expansion of its platform. This update introduces always-on background agents, a revamped investigation architecture, and a collaborative workspace for engineers and AI agents to work together on live incidents.
Multi-Agent Investigation System
The highlight of this release is a new multi-agent investigation system developed by Resolve AI’s internal research lab. Unlike previous models that utilized a single AI agent to diagnose production failures, the new system deploys a coordinated team of specialized agents. These agents can pursue multiple hypotheses simultaneously, verify each other’s findings, and construct thorough causal chains from root causes to symptoms. According to Resolve AI, this architecture has resulted in more than a twofold improvement in root cause accuracy compared to earlier versions.
Operational Context and Accuracy Claims
The announcement comes at a critical time for the software industry, where the adoption of AI-powered code generation has surged, allowing engineering teams to produce significantly more software. However, the operational aspects of maintaining that software—such as debugging and monitoring—remain largely manual. Resolve AI, which raised $125 million in a Series A round at a $1 billion valuation earlier this year, is positioning itself to capitalize on this operational gap in the software lifecycle.
While the claim of improved accuracy is notable, it is based on internal benchmarks rather than third-party audits. CEO Spiros Xanthos acknowledged the complexity of these evaluations, which reflect real-world scenarios faced by their enterprise customers, including companies like Coinbase and Salesforce.
Innovative Background Agents
In addition to incident response, Resolve AI is introducing a new class of background agents designed to manage ongoing operational tasks that engineering teams often struggle to maintain. These agents operate continuously, pre-investigating priority issues and monitoring deployments, thereby shifting teams from reactive firefighting to proactive management.
Collaborative Workspace for Engineers and AI
Another significant component of the update is a shared investigation surface, where engineers and AI agents can collaborate in real time during incidents. This workspace allows for dynamic updates and seamless interaction between human engineers and AI agents, fostering greater trust and automation in the incident response process.
Resolve AI also plans to offer its platform as a REST API, enabling integration with broader workflows and signaling its intent to be a specialized node in an ecosystem of AI agents.
Competitive Landscape and Pricing Model
As the agentic operations space becomes increasingly crowded with competitors like Datadog and PagerDuty, Xanthos emphasized Resolve AI’s unique technical foundation and its focus on domain-specific fine-tuning. The company’s pricing model is also noteworthy; it operates on an outcome-based system where credits are consumed only when agents perform actions, aligning costs directly with value delivered.
This article was produced by NeonPulse.today using human and AI-assisted editorial processes, based on publicly available information. Content may be edited for clarity and style.
