Traversal
New York City, US · 95 employees · 17 known investors
Traversal builds an agentic AI system that uses machine learning to autonomously analyze telemetry data and troubleshoot production incidents for enterprise software teams. The platform leverages causal machine learning and reinforcement learning to enable self-healing software and improve site reliability.
Also known as Traversal AI
Founders & leadership
Investors · 17
Also in the syndicate · 12
Company profile
researched Aug 2026Traversal develops an "AI SRE" (site reliability engineering) agent for enterprise production systems. The product autonomously investigates production incidents by traversing large volumes of telemetry — logs, metrics, traces, dependency relationships and code changes — to isolate root causes and propose or execute remediation, rather than surfacing symptoms on dashboards. The company frames the problem as one requiring frontier LLM/agent work: petabyte-scale noisy data, incidents that are each unique, and high per-minute cost of downtime [0][2][6].
The platform is organized around five capabilities: Alert Intelligence (autonomous triage of alerts before they become incidents), Incident RCA (tracing incidents across services, dependencies and changes), Self-healing (automated remediation), Code Resilience (feeding production context back into development to flag risky changes, e.g. warning on a pull request that a timeout change would break a downstream dependency), and Production Support (natural-language querying of live production state grounded in telemetry, code and internal knowledge). Traversal describes an internal "Production World Model™" built from customer documentation and other sources, and offers a bring-your-own-cloud (BYOC) deployment option [0][1]. Secondary coverage adds that the platform supports read-only access, on-premises deployment and bring-your-own-model configurations, requires no agents or sidecars or writes to production, generates evidence-backed remediation recommendations with confidence scores, and produces automated post-mortems [6][7].
Named deployments cited by the company include DigitalOcean, Cloudways, Eventbrite, American Express, PepsiCo, an unnamed Fortune 100 financial institution, a Fortune 50 food & beverage company, and a leading global crypto exchange [0][4][6]."]
Founding story
According to the company's about page and launch post, Traversal was started by four co-founders combining academic causal-machine-learning research with high-stakes production experience. Anish Agarwal, a professor of operations research and computer science at Columbia University with a Ph.D. in computer science from MIT (previously Boston Consulting Group, Amazon Core AI, Microsoft Research), describes leaving pure academia to commercialize his research on causal ML and synthetic data generation. Raaz Dwivedi, a professor of operations research and computer science at Cornell Tech (Ph.D. UC Berkeley; postdoc at Harvard and MIT; previously Microsoft Research and WorldQuant), applied his causal ML and reinforcement learning work to incident troubleshooting. Raj Agrawal (Ph.D. in computer science at MIT; previously Basis and Arbilex) drew an analogy between his research on how gene regulatory networks cause disease and how failures in microservices cause production incidents. Ahmed Lone (B.S. in computer science and finance, University of Pennsylvania; previously Citadel Securities) brought engineering, product and high-frequency trading experience, motivated by repeated 3 a.m. production pages [2][6].
Business model
Traversal sells its AI SRE agent to enterprise engineering organizations, with an enterprise go-to-market motion (demo booking, forward-deployed engineering, dedicated GTM sellers targeting the Fortune 500) and flexible deployment options including read-only data access, bring-your-own-cloud and on-premise installation [0][5][6][7]. The sources do not state pricing or contract structure.
Traction
The launch announcement reported deployments at DigitalOcean, Eventbrite, Cloudways and large Fortune 100 financial institutions, with greater than 90% accuracy across hundreds of high-impact incidents [6]. Customer metrics published on the website include a 70% MTTR reduction and about 96,000 support engineering hours saved per year at Cloudways (covering 845,000+ customer applications and 125,000+ annual investigations, with 95%+ accuracy on self-healing flows), a 38% MTTR reduction and 3,600 engineering hours saved annually at DigitalOcean (plus 21% fewer incidents), a 32% MTTR reduction and 82% RCA accuracy at American Express, 80% RCA accuracy and 6,000 engineering hours saved per year at a Fortune 100 customer ingesting 250 billion logs of interest per day, elimination of a backlog of 700+ high-severity alerts at PepsiCo, and projected 40%+ MTTR reduction, 75% RCA accuracy and 2,000+ senior engineering hours reclaimed per month at a global crypto exchange [0][1]. A third-party profile cites root cause analysis completed in 2-4 minutes versus hours for human teams [7].
Latest developments
The company announced "Traversal Workers," described as proactive AI SREs that act unprompted [0]. In March 2026, American Express was announced as both a partner and an investor, adopting Traversal for site reliability engineering [4]. The company blog continued publishing through August 2026 on topics including benchmarking on high-severity incidents and incident management in the age of AI [6]. Traversal is actively hiring across AI research, engineering, product, forward-deployed engineering and go-to-market roles in New York [5].
▸Full profile — market position, technology, go-to-market, geography, history
Market position
Traversal positions itself in the emerging "AI SRE" category, adjacent to and complementary with existing observability and monitoring tooling, which the company argues generates noise rather than answers and represents the second-largest software expense after cloud computing [6]. It has been named to Redpoint's Infra Red Report and 60 AI Disruptors lists, Paraform's Talent Density Index, Harmonic's Hot 25 Startups, Sequoia's "2026: This is AGI" list, and the Constellation ShortList™ for Autonomous IT Platforms [2]. LangChain CEO Harrison Chase characterizes the company as working at the front of the AI agent field with an "ambient" agent system [2].
Sources describe differentiation along several axes: agentic root cause analysis that traverses dependency graphs to find the causal "smoking gun" rather than surfacing symptoms; evidence-backed remediation with confidence scores and one-click approval; proactive alert de-noising and health checks; enterprise deployment flexibility (read-only access, on-premise/BYOC, bring-your-own-model, no agents or sidecars, no writes to production); and automated post-mortem generation [6][7]. The company also argues that its approach avoids requiring customers to maintain large amounts of hand-written context, instead constructing a Production World Model from customer systems and documentation [0][6]. Its research-heavy interdisciplinary team — professors and PhDs in causal ML alongside high-frequency trading operators and observability/DevOps product people — is presented as a core advantage [2].
Technology
The system combines an agentic LLM architecture with the founders' research in causal machine learning, allowing agents to "traverse" massive telemetry datasets and orchestrate tool calls to identify true root causes in real time [6]. Elastic founder and CTO Shay Banon states that Traversal uses Elasticsearch together with AI agents so its agents can autonomously search large-scale telemetry in enterprise settings [2]. Employees describe work on causal agentic harnesses, agent memory systems, and data processing and context engineering at petabyte scale, and the company's stack includes Python, Rust, TypeScript and Go on AWS, Kubernetes and Terraform [5]. The product builds a "Production World Model™" from customer telemetry, code and documentation, and supports BYOC deployment [0]. The company also markets "Traversal Workers," described as proactive AI SREs that act unprompted [0].
Go-to-market
The company uses a demo-request/enterprise sales motion via its website, supported by named customer case studies (DigitalOcean, Cloudways, American Express, PepsiCo) and quantified proof points on MTTR reduction and engineering hours saved [0][4]. Internally it staffs forward-deployed engineering, solutions engineering and East Coast GTM roles, and one employee notes executives at Fortune 500 accounts engaging directly; the company has also run subway advertising in New York City [5]. Customer-facing content emphasizes fast time-to-value, with one deployment cited as reaching production-ready performance in seven days [0].
Large enterprises running complex, mission-critical distributed systems — cloud and hosting providers, financial institutions, consumer/food-and-beverage multinationals, e-commerce and crypto exchanges. Named or described users include DigitalOcean, Cloudways, Eventbrite, American Express, PepsiCo, a Fortune 100 financial institution, a Fortune 50 F&B company and a leading global crypto exchange [0][4][6][7].
Geography
Headquartered in New York City, with a predominantly NYC-based, in-office team and an NYC office used for onsite interviews [5]; a third-party profile lists HQ as New York City, NY, USA [7].
History
Sequoia lists Traversal as founded in 2024 and partnered in 2024 [3]; a third-party profile likewise gives a 2024 founding date [7]. The company emerged from stealth on 2025-06-18/2025-06-20 with a combined $48 million seed and Series A led by Sequoia Capital and Kleiner Perkins [4][6]. In March 2026, press items announced that American Express partnered with and invested in Traversal to apply the technology to its site reliability engineering [4].
Compiled by commissioned research from 8 cited public sources — announcements, filings, and press listed under research sources below.
Key figures
latest reportedCompany-reported or press-reported figures, each dated to when it was claimed — not independently audited.
Competitors · 4
by search overlap
Companies competing with Traversal for the same Google search keywords, organic and paid, via search-intersection analysis.
Timeline · 4
launches, deals, and filingsPress items report that American Express engaged Traversal to transform its site reliability engineering with AI and also invested in the company.
Recognitions listed by the company include Redpoint's Infra Red Report and 60 AI Disruptors, Paraform's Talent Density Index, Harmonic's Hot 25 Startups, Sequoia's '2026: This is AGI', and the Constellation ShortList for Autonomous IT Platforms.
The company announced Traversal Workers, described as proactive AI SRE agents that act unprompted.
Traversal launched publicly, announcing combined seed and Series A rounds totaling $48 million led by Sequoia Capital and Kleiner Perkins, with participation from NFDG and Hanabi plus angel investors in infrastructure and AI.
$48M source ↗
Dated company events from announcements, filings, and press; legal rows summarize public dockets and regulator releases.
In the news
▸Research sources · 8
primary sources listed
- Traversaltraversal.com · web
8 public sources were cited for this profile; the first-party ones are listed here.
Frequently asked questions
- What does Traversal do?
- Traversal builds an agentic AI site reliability engineer that triages alerts and root-causes production incidents for enterprises.
- Who founded Traversal?
- Traversal was founded by Anish Agarwal, Raaz Dwivedi, Raj Agrawal, Ahmed Lone.
- Who are Traversal's investors?
- Traversal's investors include Kleiner Perkins, Sequoia Capital, SV Angel, Amex Ventures, Sabrina Hahn Fund.
- Where is Traversal headquartered?
- Traversal is headquartered in New York City, US.