Resolve AI
Also known as: Resolve AI, Resolve
Most heavily funded autonomous AI SRE, targeting 80 percent autonomous resolution with a multi agent parallel hypothesis team, a self learning knowledge graph, and REST plus MCP APIs.
Resolve AI is the most heavily funded of the autonomous AI SRE agents and among the most capable, built by former Splunk engineers behind OpenTelemetry and Log Insight. It reached a $1 billion valuation with more than $160 million raised, and it targets roughly 80 percent autonomous resolution of production incidents rather than mere triage. Where traditional AIOps clusters events it already has, Resolve runs an investigation: it queries logs, metrics, and traces, walks dependency graphs, reads CI and CD pipelines, and re plans each step from what it finds.
The architecture is the differentiator. Resolve runs a coordinated team of specialized agents that pursue multiple hypotheses in parallel and validate each against real evidence, which the company reports delivers more than a two times improvement in root cause accuracy over single agent approaches, all grounded in a dynamic self learning knowledge graph of the environment. It is vendor neutral, pulling from many observability and incident sources rather than requiring migration, and it ships not one agent but a suite: an incident agent that root causes and fixes, a cost optimization agent, and a feature development agent that carries production context into new work. A May 2026 update added always on background agents that audit alert hygiene, flag configuration drift, and triage on call alerts within five minutes before an engineer engages. Crucially for builders, it exposes a REST API and an MCP server, so Resolve plugs into broader agentic workflows rather than sitting as a closed endpoint. Named customers include Coinbase, which reported ten times engineering productivity and 73 percent faster time to root cause, and DoorDash, which reported up to an 87 percent reduction in time to root cause, alongside Salesforce, Zscaler, MongoDB, and Pinecone.
Resolve publishes no pricing; contracts run through enterprise sales. Its telemetry flows into the vendor cloud, the Category one tradeoff of capability for some lock in. For teams that want the deepest autonomous investigation with a knowledge graph and open API surface, Resolve sets the high water mark in the lane.
Vendor details
Canonical URL
https://resolve.ai
Category
SRE / DevOps agent
Subcategory
AI SRE agent
Funding status
Independent, founded by former Splunk engineers behind OpenTelemetry and Log Insight. Reached unicorn status at a $1 billion valuation, with a $125 million round reported in early 2026 and more than $160 million raised in total. Named customers include Coinbase, DoorDash, MSCI, Salesforce, Zscaler, MongoDB, Toast, and Pinecone.
Company status
independent
Use cases & customers
Primary use cases
Target customers
Deployment options
Integrations
Vendor neutral: pulls data from many observability and incident sources rather than owning the telemetry, and exposes a REST API plus an MCP server so it wires into broader agentic workflows. A coordinated team of specialized agents pursues multiple hypotheses in parallel and validates them against evidence, with separate agents for incidents, cost optimization, and feature development with production context.
In practice
An incident spans logs, metrics, traces, and three services and your one on call engineer is guessing. Resolve runs specialized agents on parallel hypotheses, validates each against evidence, and returns a ranked root cause.
Alerts pile up overnight before anyone looks. Resolve's always on background agents triage on call alerts within five minutes and flag configuration drift before an engineer engages.
You want the SRE agent inside your own automation, not a separate dashboard. Resolve exposes a REST API and an MCP server so it plugs into broader agentic workflows.
Sources & related URLs
Capability coverage
9.0 / 14 capabilities · 64%
| Integrations & Tool CallingVendor neutral, pulls from many observability and incident sources; queries logs, metrics, traces, and cloud APIs during investigation, Resolve docs 2026-07-06 | Full |
|---|---|
| Workflow OrchestrationA coordinated team of specialized agents pursues multiple hypotheses in parallel and validates against evidence, a flagship multi agent orchestration, Resolve docs 2026-07-06 | Full |
| Knowledge Grounding & RAGGrounds investigations in a dynamic self learning knowledge graph of the environment, a flagship grounding architecture, Resolve docs 2026-07-06 | Full |
| Human Oversight & GuardrailsTargets roughly 80 percent autonomous resolution with engineer oversight, but pushes further into autonomous fixes than the read only investigators, Resolve docs 2026-07-06 | Partial |
| Security, Identity & GovernanceEnterprise security posture serving Coinbase, Salesforce, and Zscaler; access to production observability under enterprise governance, Resolve docs 2026-07-06 | Full |
| Observability & AuditabilityEvidence based hypothesis validation and auto generated post mortems provide investigation transparency; per step audit console less documented, Resolve docs 2026-07-06 | Partial |
| Memory & State PersistenceSelf learning knowledge graph persists and updates environmental understanding across incidents; framed as grounding rather than a distinct memory product, Resolve docs 2026-07-06 | Partial |
| Deployment & Data ResidencySaaS with telemetry flowing into the vendor cloud, the Category one tradeoff; no self host or residency option documented, Resolve docs 2026-07-06 | Partial |
| Prebuilt Agents, Templates & PacksShips a suite of prebuilt agents: incident root cause and fix, cost optimization, feature development, plus always on background agents, Resolve docs 2026-07-06 | Full |
| Triggers & Channel CoverageAlways on background agents triage on call alerts within five minutes plus alert driven investigation around the clock, Resolve docs 2026-07-06 | Full |
| Model Flexibility & RoutingProprietary agent pipeline; no customer facing model choice or routing documented, Resolve docs 2026-07-06 | Unable to verify |
| APIs, SDKs & MCP ExtensibilityExposes a REST API and an MCP server so Resolve wires into broader agentic workflows, a first class extensibility surface, Resolve docs 2026-07-06 | Full |
| Testing, Debugging & OptimizationInternal evaluation sets referenced but no customer facing testing or evaluation tooling documented, Resolve docs 2026-07-06 | Unable to verify |
| Browser & Computer UseOperates through tool and API calls, not browser or computer interface control, Resolve docs 2026-07-06 | Unable to verify |
Pricing
Contact sales; enterprise contracts, no public rates
enterprise contract
Included quota
Platform contract covering autonomous investigation with the multi agent team and knowledge graph; additional agents scoped as needed. No public tiers.
What is public
Nothing numeric; the REST API and MCP server and the three agent lines are public.
Billing mechanics
Enterprise platform contracts through sales, likely scaling with environment size and the number of agent lines deployed. Pricing not disclosed.
Cost watchouts
Three agent lines (incident, cost optimization, feature development) may each carry cost, so a full deployment scales beyond a single incident agent. Telemetry flows into the vendor cloud, the Category one lock in tradeoff.
Variable cost rationale
Enterprise platform licensing; no usage metering documented, though adding agent lines raises cost. Exposure sits at low absent published usage billing.
Additional watchouts
Most capable in the lane but Category one architecture means telemetry lives in the vendor cloud; weigh lock in against depth. Multiple agents can expand the contract.
Overage / add-ons
No public metering documented.
Sales call required
Yes — required for paid access
Free / trial
Enterprise evaluations and proofs of value through sales; no self serve trial
Lowest paid plan
None public; enterprise contract only
Commercial notes
Independent, $1 billion valuation, more than $160 million raised, from founders behind OpenTelemetry and Log Insight. Coinbase reported ten times productivity and 73 percent faster RCA; DoorDash reported up to 87 percent reduction in time to RCA.
Key ambiguities
Nothing numeric is public, and whether the cost optimization and feature development agents are bundled or priced separately is not documented.
Related vendors
- Causely — Causal reasoning engine that infers the single root cause of an…
- Cleric — Safety first AI SRE that investigates alerts read only and delivers…
- incident.io — All in one incident management platform in Slack and Teams with a…
- NeuBird — Autonomous AI SRE built on an Agentic Context Engine that reasons…
- Rootly — AI native incident management platform whose AI SRE runs parallel…
- Traversal — Enterprise AI SRE built on a causal Production World Model and…