Resolve AI Expands Platform with Always-on Agents and New Investigation Architecture

New platform capabilities include autonomous background agents, collaborative investigation workflows, and a new architecture delivering more than 2x improvement in root cause accuracy

SAN FRANCISCO, May 21, 2026 /PRNewswire/ -- AI agents are rapidly changing how software gets built. But operating that software in production still depends heavily on manual work from engineering teams. Resolve AI is built to help organizations operate production systems at AI speed. Today, the company announced a major expansion of its AI for prod platform, introducing always-on background agents that continuously perform operational work in production, a new investigation architecture that improves incident investigation quality, and collaborative spaces where engineers and AI agents work from the same evidence in real-time.

Many of the operational tasks required to keep production systems healthy still happen reactively and manually. Teams are expected to continuously monitor deployments, investigate alerts, audit operational hygiene, track changes across complex environments, and handle an ever-growing list of daily tasks. These responsibilities are critical, but difficult to sustain consistently at scale. Resolve AI's new always-on background agent layer is designed to handle those ongoing operational demands. When engineers open Resolve AI, agents have already been working: pre-investigating priority issues, monitoring deployments, auditing alert hygiene, flagging configuration drift, and surfacing cost anomalies. Instead of starting investigations from scratch, teams begin with verified findings and recommended next steps. Agents can run continuously on schedules or wake automatically in response to events such as deploys and alerts, while accumulating knowledge from every investigation and interaction over time.

"Engineers are not limited just by skill; they are limited by time and context," said Mayank Agarwal, Founder and CTO of Resolve AI. "Running production well requires understanding your systems more deeply than any on-call rotation allows, and that is what we built Resolve AI to change. With always-on agents handling daily production work, and improved investigation quality, engineers can finally focus on the problems that actually move the product forward."

Alongside the continuous operational work, incident investigation is materially stronger. Resolve AI agents now act as first responders for every on-call alert, typically triaging within five minutes before an engineer is involved. For more complex issues, the company has introduced a new investigative architecture developed by its AI Lab that delivers more than 2x improvement in root cause accuracy on internal evaluation sets compared to earlier versions of the platform – the same system that helped customers like DoorDash reduce time to root cause by up to 87%.  The architecture uses a coordinated team of specialized agents to investigate multiple hypotheses and evidence sources in parallel, independently verifying conclusions along the way. Leading enterprises, including Coinbase, MSCI, Salesforce, and Zscaler, rely on Resolve AI to manage their production environments.

When engineers engage in an active incident, they work alongside agents and peers within a shared investigation surface built around the same live evidence and context. Reports update dynamically as investigations evolve, every finding is inspectable, and engineers can explore side investigations without interrupting the primary workflow. Evidence is embedded directly into the workspace, source queries are pullable and modifiable in place, and remediation can be triggered from the same interface without context switching. Resolve AI is now also accessible as a REST API and an MCP server, allowing engineering teams to integrate the platform into broader agentic workflows and infrastructure.

For engineering teams, today's announcements change what it means to run production. Reliability is no longer a function of how many engineers are available to respond. Operational work that required constant human attention, from monitoring deployments to auditing alert hygiene to following up on incidents, now runs continuously without anyone initiating it. The engineers who used to spend their time inside every production problem can now set the standards for how production should behave and trust that agents are managing against them.

About Resolve AI
Resolve AI is AI for running and operating software in production, which underpins how businesses operate. Founded by observability pioneers Spiros Xanthos and Mayank Agarwal, Resolve AI combines custom AI models, production-specific agents, and deep systems expertise to solve the hardest problems in modern software operations. Leading companies, including Coinbase, DoorDash, MSCI, Salesforce, and Zscaler, rely on Resolve AI to run and operate their production environments. Learn more at resolve.ai.

SOURCE Resolve AI