Resolve.ai
Traversal
Most aggressive AI SRE with $1B valuation, Splunk pedigree, and goal of 80% autonomous incident resolution
Highest accuracy claim (90%+) with proven results: DigitalOcean saved 36,000 engineering hours annually
- 80% auto-resolve goal
- $1B valuation
- Splunk founders
- Coinbase 10x boost
- 90%+ accuracy
- On-prem option
- 36K hrs saved
- No public pricing
- SOC2 not confirmed
- No auto-remediation
- SOC2 not confirmed
Key Takeaways
- Resolve.ai — $1B unicorn (Dec 2025) targeting 80% autonomous resolution. Founded by Splunk architects. Best for Fortune 500 scale.
- Cleric — Gartner Cool Vendor 2025. Self-learning agent with read-only safety approach. Best for mid-market SaaS on Kubernetes.
- Traversal — Academic ML pedigree with 90%+ accuracy claim. DigitalOcean saved 36,000 engineering hours/year. Best for outcome-focused teams.
- Bottom Line — Choose Resolve.ai for aggressive autonomy, Cleric for safety-first approach, Traversal for proven accuracy at scale.
Overview: The Rise of AI SRE Agents
The AI SRE market exploded in late 2025, with Resolve.ai achieving $1B unicorn status just weeks after launching. Unlike traditional AIOps tools that correlate alerts, these new AI SRE agents actively investigate incidents—forming hypotheses, querying systems, and (increasingly) fixing issues autonomously.
Resolve.ai takes the most aggressive approach, targeting 80% autonomous resolution with founders from Splunk's core observability team. Cleric prioritizes safety with a read-only approach that learns from every investigation. Traversal brings academic rigor with causal ML expertise, claiming 90%+ accuracy in root cause analysis.
This comparison helps you choose based on your organization's risk tolerance, scale requirements, and incident response maturity.
Feature Comparison
Here's how Cleric, Resolve.ai, and Traversal compare across core capabilities, integrations, and security considerations.
Feature Matrix
| Feature | Resolve.ai | Cleric | Traversal |
|---|---|---|---|
| Core Capabilities | |||
| Root Cause Analysis | Real-time event correlation | 5-min avg diagnosis, linked evidence | 2-4 min, 90%+ accuracy |
| Auto-Remediation | Target: 80% autonomous resolution | Read-only, supervised (roadmap) | Recommendations + health checks |
| Self-Learning | Dynamic knowledge graph | Improves with every interaction | Causal ML from incidents |
| 24/7 Monitoring | Continuous alert monitoring | Autonomous alert investigation | Ambient anomaly detection |
| Integrations | |||
| Cloud Providers | AWS | AWS, GCP, Azure | Cloud-agnostic |
| Observability Tools | Joins across tools | Datadog, Grafana, Prometheus, Elastic | Grafana, Elastic, VictoriaMetrics |
| Incident Platforms | Slack | PagerDuty, Slack | Slack, Alertmanager |
| Code & Infra | GitHub, Kubernetes | GitHub, Kubernetes, Confluence | GitHub, Confluence |
| Security & Compliance | |||
| SOC2 Certification | Enterprise-grade (not confirmed) | Regular pen testing | Not publicly confirmed |
| Data Privacy | Enterprise security | Never trains on customer data | Read-only access option |
| On-Premise Option | Not confirmed | | On-prem deployment available |
| Access Controls | Enterprise RBAC | SSO/SAML, RBAC, audit logs | Read-only by default |
| Company & Funding | |||
| Valuation | $1B (Dec 2025) | $9.8M seed stage | $48M Seed+A |
| Lead Investors | Lightspeed, Khosla, Greylock | Vertex Ventures, Zetta | Sequoia, Kleiner Perkins |
| Founders | Ex-Splunk (OpenTelemetry, Log Insight) | Stealth team | Columbia, Cornell, MIT professors |
| Notable Customers | Coinbase, 100+ F500 committed | BlaBlaCar | DigitalOcean, Eventbrite |
Resolve.ai: The $1B Unicorn
Resolve.ai made headlines in December 2025 when it raised $250M at a $1B valuation—making it the fastest AI SRE startup to achieve unicorn status. Founded by ex-Splunk executives who helped create OpenTelemetry and VMware's Log Insight, the team brings unparalleled observability pedigree.
The company's ambitious goal: auto-resolve 80% of production incidents without human intervention. While current ARR is reportedly ~$4M, the pipeline of 100+ Fortune 500 commitments suggests rapid enterprise adoption ahead.
Resolve.ai
- Most aggressive autonomy goal: 80% incidents resolved without humans
- Fastest to unicorn status in AI SRE category ($1B in Dec 2025)
- Founded by Splunk architects who created OpenTelemetry
- 100+ Fortune 500 companies in committed pipeline
- Coinbase reported 10x engineering productivity boost
- No public pricing (enterprise-focused, likely expensive)
- SOC2 certification not publicly confirmed
- Younger product, less proven at scale vs. incumbents
- Limited public documentation on integrations
Cleric: The Safety-First Agent
Cleric takes a fundamentally different approach: it's designed to be read-only. The agent observes, investigates, and recommends—but requires human approval before any action is taken. This conservative posture earned Cleric Gartner Cool Vendor 2025 recognition in AI for SRE and Observability.
The self-learning system improves with every investigation and engineer interaction, providing confidence scores and linked evidence for transparency. For organizations prioritizing safety over speed, Cleric offers peace of mind.
Cleric
- Gartner Cool Vendor 2025 recognition
- Self-learning system that improves signal-to-noise ratio over time
- Conservative read-only approach prioritizes safety
- Transparent reasoning with linked evidence for every finding
- Never trains on proprietary customer data
- Read-only means no automated fixes (human-in-the-loop required)
- Smaller funding ($9.8M) limits R&D velocity
- Auto-remediation still on roadmap, not available today
- Less brand recognition than competitors
Traversal: Academic ML Expertise
Traversal was founded by Columbia and Cornell professors specializing in causal machine learning. This academic rigor translates to the market's highest accuracy claim: 90%+ accuracy in 2-4 minutes for root cause analysis.
The DigitalOcean case study provides compelling validation: 38% MTTR reduction, 36,000 engineering hours saved annually, with 50+ engineers actively using the platform. Traversal also offers on-premise deployment—a key differentiator for regulated industries.
Traversal
- Highest accuracy claim in market: 90%+ in 2-4 minutes
- Academic ML expertise from Columbia, Cornell, MIT founders
- DigitalOcean case study: 38% MTTR reduction, 36K hours saved/year
- On-premise deployment option for regulated industries
- Outcome-driven pricing (not data-volume based)
- SOC2 status not publicly confirmed
- Smaller customer base (newer to market)
- Focused recommendations vs. automated remediation
- Pricing not publicly available
Market Context: AI SRE in 2026
The AI SRE market represents a significant evolution from traditional AIOps. Key differences:
- AIOps (2017-2023): Rule-based correlation, noise reduction, pattern matching. Tells you something is wrong.
- AI SRE (2024+): LLM-powered investigation, hypothesis formation, causal analysis. Tells you why and how to fix it.
The incumbents are responding: Datadog launched Bits AI SRE (GA December 2025), PagerDuty has AI-assisted features, and New Relic is investing heavily in AI. But the pure-play startups (Resolve.ai, Cleric, Traversal) are moving faster on true autonomous capabilities.
Who Should Use What?
Based on your organization's priorities and constraints:
Best for aggressive autonomy at scale
- You're a Fortune 500 with complex production environments
- Maximum autonomy (80% auto-resolution) is your goal
- You want tools built by proven observability leaders
- Budget is secondary to reducing on-call toil
- You need to scale incident response across large teams
Best for safety-first self-learning
- You prefer safety over speed (read-only approach)
- Self-learning that improves over time appeals to you
- You run Kubernetes and microservices architecture
- Gartner recognition gives you confidence for procurement
- You want transparent reasoning you can verify
Best for proven accuracy + on-prem
- Accuracy is more important than speed of automation
- You need on-premise deployment for compliance
- Your team values evidence-based recommendations
- You're similar to DigitalOcean scale (50+ engineers)
- You prefer outcome-based vs. data-volume pricing
Enterprise & Compliance Considerations
For enterprise deployments, compliance posture varies significantly across these newer entrants:
- Resolve.ai: Enterprise-grade security (Fortune 500 commitments suggest robust controls), but SOC2 not publicly confirmed. Greylock/Khosla/Lightspeed backing provides credibility.
- Cleric: Regular third-party penetration testing, SSO/SAML, RBAC, audit logs. Never trains on customer data. Read-only access minimizes blast radius.
- Traversal: On-premise deployment option with read-only access. SOC2 not publicly confirmed. Sequoia/Kleiner Perkins backing.
For regulated industries (healthcare, finance), consider pairing these tools with established platforms: Datadog Bits AI (HIPAA-capable), Rootly (SOC2 since 2022), or incident.io (SOC2 + ISO 27001).
Integration Ecosystem
All three tools integrate with core observability and incident management tools:
- Common: Slack, GitHub, Kubernetes, basic observability stacks
- Cleric advantage: Broadest observability coverage (Datadog, Grafana, Prometheus, Elasticsearch) plus Confluence for documentation context
- Traversal advantage: Deep integration with Grafana, Elastic, VictoriaMetrics; parses codebase alongside telemetry
- Resolve.ai approach: "Joins data across observability tools, infrastructure, code, and release pipelines"—suggesting broad connectivity
Frequently Asked Questions
Which AI SRE tool has the best MTTR reduction?
Resolve.ai claims up to 80% MTTR reduction with their goal of 80% autonomous resolution. Datadog Bits AI reports 70% MTTR reduction. Traversal showed 38% MTTR reduction at DigitalOcean (third-party validated). Rootly claims 81% MTTR reduction. The key is understanding that these metrics depend heavily on baseline maturity—immature teams see larger percentage improvements.
Can these AI SRE tools actually fix production issues automatically?
Today, most are cautious: Cleric is read-only by design (observes and recommends). Traversal provides recommendations but doesn't execute fixes. Resolve.ai has the most aggressive posture, targeting 80% autonomous resolution, but specific auto-remediation capabilities aren't publicly detailed. Datadog Bits AI proposes code fixes. The industry is moving toward more autonomy, but with careful human-in-the-loop controls.
Are these tools SOC2 certified?
SOC2 status varies: Datadog (Bits AI parent) is SOC2 Type II certified. Rootly has been SOC2 Type II since January 2022. incident.io is SOC2 Type II and ISO 27001 certified. For the pure-play AI SRE agents: Cleric notes regular penetration testing but no public SOC2 claim. Resolve.ai and Traversal don't have publicly confirmed SOC2 certifications yet.
How do AI SRE tools differ from traditional AIOps?
Traditional AIOps (Moogsoft, BigPanda, etc.) focus on alert correlation and noise reduction with rule-based systems. AI SRE tools like Resolve.ai, Cleric, and Traversal use LLMs and causal ML to actually understand what's happening—forming hypotheses, running queries, and proposing (or executing) fixes. The key difference: AIOps tells you something is wrong; AI SRE tells you why and how to fix it.
Which tool is best for regulated industries (healthcare, finance)?
For regulated industries: Traversal offers on-premise deployment with read-only access. Datadog Bits AI supports HIPAA. Rootly partners with Secureframe for SOC2, ISO 27001, PCI DSS, HIPAA, and GDPR compliance. incident.io has SOC2 Type II and ISO 27001. The pure-play AI SRE startups (Resolve.ai, Cleric, Traversal) are newer and may need more due diligence for regulated environments.
What's the typical pricing for AI SRE tools?
Most AI SRE tools don't publish pricing: Resolve.ai, Cleric, and Traversal are enterprise-focused with custom quotes. Datadog Bits AI uses per-investigation pricing on top of Datadog subscriptions. Rootly starts around $240/user/year (Essentials). incident.io has a free tier (5 users), with paid plans from $19-50/user/month. Generally, expect $10K-$100K+ annually for enterprise deployments.
Final Verdict
The right AI SRE tool depends on your organization's risk tolerance and scale:
- Choose Resolve.ai if you're a Fortune 500 seeking maximum autonomy and can afford enterprise pricing. The Splunk pedigree and aggressive 80% auto-resolution target make it the most ambitious player.
- Choose Cleric if safety is paramount. The read-only approach, self-learning capabilities, and Gartner recognition make it ideal for organizations that want AI assistance without giving up control.
- Choose Traversal if proven accuracy matters most. The DigitalOcean case study (36K hours saved) and on-premise option make it compelling for mid-market to large enterprises with compliance requirements.
For organizations not ready for pure-play AI SRE startups, consider Datadog Bits AI (if you're already a Datadog customer) or incident.io (for Slack-native incident management with AI capabilities). The market is moving fast—expect significant feature parity improvements across all players in 2026.
Ready to Find the Right AI Tools?
Browse our data-driven rankings to find the best AI tools for your team.