Traversal
AI SRE agents that join the incident channel and trace the root cause
Overview
Traversal is an enterprise AI SRE platform from New York that maps a production environment into a live dependency model and uses causal search to find the root cause of incidents. Its Workers agents join Slack or Microsoft Teams channels on their own to investigate incidents and triage alerts. It is sold by quote after a scoping call.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 14 Oct 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
- Incident RCA and Alert Intelligence
- Self-healing and Code Resilience capabilities
- Production Support
Use cases
What you can produce with Traversal
- A root-cause analysis for a production incident, traced across services, dependencies and recent changes
- Incident Workers that join a Slack or Microsoft Teams incident channel and investigate unprompted
- Alert Workers that dedupe noisy alert channels and surface the novel alerts with reasoning attached
- A live Production World Model of services and dependencies built from existing telemetry and code
- A scratchpad per incident that becomes the postmortem draft
- Read-only, agentless connections to Datadog, Splunk, Grafana, Dynatrace, GitHub, PagerDuty and others
- Traversal context inside AI coding tools through its MCP server
ASEAN Perspective
Traversal in Southeast Asia
Traversal has not announced any APAC customers, regional offices, or Southeast Asia partnerships as of mid-2026; all confirmed deployments are at North American enterprises. For ASEAN enterprises — particularly large banks, telecoms, and e-commerce platforms in Singapore, Indonesia, Malaysia, and the Philippines — the technology is relevant given the region's rapid shift to complex microservices architecture, but procurement will require engaging the US-based sales team and evaluating data residency implications. On-premises deployment support may help regulated ASEAN financial institutions (e.g., MAS-regulated entities in Singapore) meet local data sovereignty requirements, but this must be confirmed directly with Traversal.
Traversal is one of the more serious AI SRE products: a causal-inference research team, a read-only, agentless deployment over an existing observability stack, and named enterprise customers including American Express, PepsiCo and DigitalOcean. The September 2026 general availability of Incident Workers, which join an incident channel and investigate unprompted, makes it a teammate rather than a dashboard.
Every performance figure is the vendor's, and its own MTTR averages range from 43% to over 85%. Pricing is by quote only, with no trial, and support is centred on New York. Best for large enterprises with a mature stack and heavy alert noise.
What we found
Traversal is a New York company building what it calls an AI SRE: software that investigates production incidents by reading an organisation's logs, metrics, traces, code and deploys. Its four founders come from causal machine learning research and trading: Anish Agarwal is a Columbia professor, Raaz Dwivedi teaches at Cornell Tech, Raj Agrawal did a doctorate at MIT, and Ahmed Lone traded at Citadel Securities. The company dates itself to 2023 and left stealth in June 2025 with $48 million across a seed round led by Sequoia and a Series A led by Kleiner Perkins. In March 2026 American Express became a customer and Amex Ventures invested; SiliconANGLE put that investment at $5 million and total funding at about $53 million.
The system builds a Production World Model of services and dependencies and runs a Causal Search Engine over it to converge on a root cause. It connects read-only, without installed agents, to tools such as Datadog, Splunk, Grafana, Dynatrace, GitHub and PagerDuty, and Traversal says raw telemetry stays in the customer's environment. Its 2026 releases moved it into chat: an MCP server in March, Agentless Data Capture in July, Microsoft Teams support in August, then on 2 September general availability of Incident Workers, agents that join an incident channel unprompted, with Alert Workers in public beta. Knowledge Bank 2.0, which folds in runbooks and past incidents, followed in September.
Treat the results as vendor figures. The homepage credits American Express with a 32% cut in resolution time, Cloudways 70% and PepsiCo with clearing 700+ high-severity alerts; DigitalOcean reports 38%. An unnamed Fortune 100 financial firm reports 82% root-cause accuracy and a Fortune 50 food and beverage company 80%. Traversal's own averages disagree: 43% on its customers page, over 85% in a summary file on its site.
There is no published price, trial or self-serve plan; pricing follows a scoping call. The team is predominantly in New York, and no APAC office or customer was found.
Compiled by RECATOOLS from public sources: vendor documentation and pricing, code repositories and published reporting.
Notable facts
- The causal ML techniques powering Traversal were originally developed for learning gene regulatory networks — the founders adapted methods for tracing CRISPR interventions in biology to trace code changes in production systems.
- Traversal processes over 250 billion logs of interest daily for American Express across Kubernetes, Lambda, mainframe, and on-premises environments — all to find a single root cause.
- The founding team hit 0% accuracy on their first enterprise pilot when they scaled from small companies to those with thousands of microservices, which they celebrated with a Negroni before rebuilding the architecture from scratch.
- Raaz Dwivedi, one of the co-founders, ranked All India Rank 10 in the IIT-JEE entrance exam — one of the world's most competitive engineering tests.
Frequently asked questions
About this listing
This entry was compiled from publicly available data including Traversal's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Traversal unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Traversal directly →
Spotted something out of date? Suggest an update →
Traversal in the news
Cybersecurity
Unit 42 Links Fortinet Path-Traversal Bug to Memory-Only FortiGate Implant in Telecom Intr...
Cybersecurity
Adobe ColdFusion CVE-2026-48282: A Perfect-Score Path-Traversal Flaw CISA Says Is Being Ex...
Business
Ahold Delhaize Notified Regulators of Its Cyber Incident. Dutch Police Say It Has Not File...
More in Code & Dev Tools
