Resolve AI
by Resolve AI
AI agents that take production on-call, investigate incidents across code and telemetry, and act inside your guardrails
Resolve AI is an agentic platform for software production operations: its agents triage alerts, investigate incidents in parallel across code, infrastructure and telemetry, and execute governed remediation such as silencing an alert, reverting a commit or opening a pull request. It is built for engineering organisations whose on-call burden and mean time to recovery have outgrown the humans carrying the pager.
Resolve AI was founded in 2024 by Spiros Xanthos, who co-created OpenTelemetry, founded Log Insight (acquired by VMware) and Omnition (acquired by Splunk), and was SVP and GM of Observability at Splunk, together with Mayank Agarwal. The product continuously pulls context across code, observability data, deployments, cloud infrastructure, configuration and operational history into a queryable knowledge graph of services, dependencies, deploys and team ownership, and then runs agents against it. Four agent types ship: on-call agents that triage alerts, suppress noise and route to the right team; an incidents agent that investigates across code, infrastructure and telemetry in parallel to produce a root cause with supporting evidence; background agents that watch deployments, generate operational reports and optimise resources; and custom agents customers build on the same primitives. Actions are governed rather than open-ended — silencing an alert, reverting a commit, opening a pull request or running a workflow, each configurable as autonomous or approval-required, with permissions scoped at organisation, team and individual level. It pairs frontier models with domain-specialised models and exposes itself over MCP, an API and a Skills framework, with more than 60 pre-built integrations across repositories, infrastructure, observability, incident management and CI/CD. Named customers include Coinbase, DoorDash, MongoDB, MSCI, Salesforce, Zscaler, Snowflake, Robinhood, Autodesk, Toast and Expedia Group. The company raised a $125M Series A at a $1B valuation led by Lightspeed in 2025, then a $40M extension at $1.5B co-led by DST Global and Salesforce Ventures announced 16 April 2026, taking total funding past $190M; the same announcement launched Resolve AI Labs, a research group building operations-specific models and evaluation frameworks, with Dhruv Mahajan, formerly of Meta's Llama post-training team, as Chief AI Scientist.
A VP of Engineering or Head of SRE at a company already running mature observability — structured logs, distributed tracing, indexed runbooks — whose on-call rotation is the bottleneck and who can fund a six-figure, sales-led platform purchase.
Incidents get a root cause with supporting evidence assembled automatically across code, deploys and telemetry before a human joins the call, so fewer engineers get pulled into war rooms.
At a Glance
- Category
- AI Agents & Orchestration
- Pricing
- Contact for pricing
- Target Market
- CTOs, VP Engineering, Site Reliability Engineers, Enterprise Developers
- Deployment
- Cloud-only, API-based
- Founded
- 2024
- Headquarters
- San Francisco, United States
- Team Size
- 11-50
Key Features
- ✓Production knowledge graph
A continuously updated, queryable graph of services, dependencies, deploys and team ownership that gives every agent the same shared model of how the estate actually fits together.
- ✓On-call agents
Triage incoming alerts, suppress known noise and route the genuine ones to the owning team, so the pager fires for signal rather than volume.
- ✓Incidents agent
Investigates an incident in parallel across code, infrastructure and telemetry and returns a root cause with the evidence it used, instead of a summary you have to re-verify.
- ✓Governed actions with scoped permissions
Silencing alerts, reverting commits, opening pull requests and running workflows are each configurable as autonomous or approval-required, scoped at organisation, team and individual level.
- ✓Background agents
Watch deployments for regressions, produce operational reports and flag resource optimisation opportunities without anyone opening a ticket.
- ✓60+ pre-built integrations
Connectors across code repositories, cloud infrastructure, observability platforms, incident management and CI/CD, so the context assembly does not require a custom integration project.
- ✓MCP, API and Skills interfaces
Exposes its capabilities to other agents and internal tooling over Model Context Protocol and a public API, and lets teams package their own runbooks as Skills.
Capabilities
Use Cases
- •Reducing war-room headcount on revenue-critical incidents
DoorDash reports pulling fewer engineers into war rooms for its ads platform, which it ties directly to advertiser trust and revenue protection.
- •First-pass alert triage overnight
On-call agents absorb the alert volume outside business hours, suppress recurring noise and escalate only what a human genuinely needs to see.
- •Cross-tool root cause analysis
When a latency spike follows a deploy and a config change, the incidents agent correlates all three rather than leaving an engineer to tab between dashboards.
- •Post-incident reporting and deploy monitoring
Background agents watch new deployments for regressions and generate the operational reports that otherwise consume a senior engineer's week.
- •Codifying runbooks as executable Skills
Existing manual runbooks become agent-executable Skills with scoped permissions, so institutional knowledge survives team turnover.
Ideal For
Best For
- ✓Reducing the number of engineers pulled into production incidents at companies with large, dependency-heavy service estates
- ✓Automated first-pass alert triage and noise suppression ahead of a human on-call rotation
- ✓Root-cause investigation that has to correlate a deploy, a config change and a telemetry signal across separate tools
- ✓Governed auto-remediation where each action class is explicitly scoped as autonomous or approval-required
- ✓Teams standardising production knowledge into a queryable service and ownership graph rather than tribal memory
Not Ideal For
- ✗Teams without mature observability — the platform reasons over logs, traces, deploys and runbooks, so a thin telemetry estate gives the agents little to work with
- ✗Organisations that need self-serve evaluation: there is no published pricing, no free tier and no trial, and buying starts with a call to their team about deployment and integration planning
- ✗Small engineering teams whose on-call load does not justify an enterprise contract that is quoted rather than listed
- ✗Buyers who require full autonomy out of the box — the value depends on carefully tiering which actions agents may take without approval
Integrations
Deployment
Market Analysis
Pros
- ✓An unusually strong named-customer list for a company this young — Coinbase, DoorDash, MongoDB, MSCI, Salesforce, Zscaler, Snowflake, Robinhood, Autodesk, Toast and Expedia Group
- ✓Investor validation from operators who are also customers: the Series A extension was co-led by Salesforce Ventures while Salesforce runs the product
- ✓Governance model is granular — autonomous versus approval-required per action class, with permissions scoped at organisation, team and individual level
- ✓SAML SSO, RBAC, data redaction and encryption, audit trails and severity-based vulnerability SLAs are documented on the product page
- ✓60+ pre-built integrations plus MCP and API access reduce the integration work that usually sinks AIOps pilots
Cons
- ✗Compliance posture is described as SOC 2, GDPR and HIPAA 'aligned' rather than certified, and no audit report or certificate is published on the site
- ✗Zero pricing transparency: no tiers, no metering unit, no trial, no free tier — every evaluation starts with a sales call
- ✗No independent review corpus exists. G2 listings for 'Rezolve.ai' and 'Resolve Systems' are different companies, so there is no neutral rating to check against the vendor's claims
- ✗A competitor-published review roundup and the vendor's own contact-sales pricing page both point to a weeks-long, solutions-engineering-led onboarding rather than self-serve adoption
- ✗Value depends on the customer already having structured logs, distributed tracing and indexed runbooks; teams without that foundation are buying an agent with nothing to reason over
- ✗The headline metrics on the site — up to 5x faster MTTR, 75% productivity gains — are vendor-reported with no published methodology
Pricing
Enterprise
Contact for pricing
- ✓All agent types and the production knowledge graph
- ✓60+ pre-built integrations
- ✓SAML SSO, RBAC, data redaction and encryption, audit logging
- ✓Deployment guidance and integration planning included in the sales process
No list pricing is published at any tier — the pricing page is a contact form offering a conversation about pricing, deployment guidance and integration planning for on-call, incidents and background agents. There is no free tier, no self-serve trial and no published metering unit, so evaluation requires a sales cycle and onboarding is a solutions-engineering-assisted project rather than a signup.
Security & Compliance
Sources
This page was written from 7 sources, 4 on domains other than resolve.ai.
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
Sema4.ai
Enterprise AI agents for knowledge work, running natively in your Snowflake or cloud account
Sapiom
Agent infrastructure that routes, runs and meters AI agents in production
Itential FlowAI
Governed AI agents for network and infrastructure operations, with deterministic execution and full audit trails
OpenAI Presence
Deploy production-grade AI voice and chat agents with enterprise policies, guardrails and evals