Harmony Raises $34M Seed for AI Enterprise Service

Harmony has raised $34M in seed funding to build an AI-native enterprise service management platform that resolves employee requests inside Slack and Microsoft Teams. Lightspeed Venture Partners led the round, with Hitachi Ventures, Fin Capital, Mercer Ventures, and Operator Partners participating.

The July 28, 2026 announcement puts significant early capital behind a simple but consequential idea: enterprise service software should complete more work, not merely route tickets to people. Harmony is applying AI agents to the repetitive requests that move through IT, HR, finance, procurement, legal, and security, where delays are measurable and employee patience is not an infinite resource.

What Happened

The $34M seed round also includes angel investors Assaf Rappaport, co-founder and CEO of Wiz, and Ofir Ehrlich, co-founder and CEO of Eon.io. Harmony said it will use the funding to accelerate product development, expand its team, deepen integrations across enterprise software, and broaden its capabilities across infrastructure and operations, enterprise applications, cybersecurity, and ERP.

Harmony was founded in 2025 by Nitzan Shapira, CEO, and Ran Ribenzaft, CTO. The pair previously built cloud observability company Epsagon, which Cisco acquired in 2021. Harmony and Calcalist have described that transaction as a $500M deal, while Cisco's acquisition page confirms the acquisition without disclosing the purchase price.

The founders' history matters because this is not a first-time team learning enterprise software by reading analyst reports. Shapira and Ribenzaft have already built, funded, and sold a cloud infrastructure company, then spent years inside Cisco watching large organizations operate. That does not guarantee Harmony will win, but it helps explain why Lightspeed is backing a large seed round in a category that is both operationally difficult and commercially mature.

Why Enterprise Service Management Is Ready for AI

Enterprise service management is often experienced as organizational archaeology. An employee needs application access, a replacement device, payroll guidance, a policy answer, or onboarding support, then must discover which portal, form, queue, or team owns the request. Each system may work as designed while the overall experience still wastes time.

Traditional service platforms improved how organizations catalog, route, and track that work. AI changes the question from "Who should receive this ticket?" to "Can the software safely resolve this request?" That distinction is Harmony's market wedge, and it is more meaningful than another chatbot attached to a legacy queue.

Harmony's agents operate in Slack and Microsoft Teams, where employees already communicate. The platform uses an organizational context graph connecting an employee's identity, devices, applications, permissions, and work history, then acts across integrated enterprise systems. The goal is to make the complexity behind a request invisible without forcing companies to replace every existing system at once.

The Product Wedge Is Measurable Work

Harmony says it offers more than 100 prebuilt agents and serves customers including n8n, eToro, Cyera, and KITH. According to the company, deployments have deflected 48% of support requests within two weeks and more than 75% within three months. Some organizations have moved beyond 90% autonomous resolution as they expanded usage, but those figures remain company-reported rather than independently audited benchmarks.

Those metrics are still strategically revealing because they focus on completed work. Enterprise AI pitches often lead with model sophistication, demo quality, or the number of workflows a platform could theoretically automate. Internal service teams care about resolution time, request volume, employee adoption, system accuracy, and how much expert capacity is returned to higher-value problems.

Harmony's early expansion beyond IT is another important signal. The company reports customers using its agents across HR, procurement, DevOps, security, and finance, supporting the broader enterprise service management thesis. If the same context and action layer can resolve requests across departments, Harmony can grow from a help-desk wedge into infrastructure for internal operations.

What the Investor Group Is Betting On

Lightspeed is not simply betting that companies want more AI software. The sharper thesis is that large organizations have accumulated hundreds of applications and workflows while employee service remains fragmented. A platform that understands intent, retrieves the right organizational context, and takes approved action could reduce both the employee's search cost and the support team's repetitive workload.

The participating investors add several useful perspectives. Hitachi Ventures brings experience with large industrial and enterprise environments, while Fin Capital, Mercer Ventures, and Operator Partners broaden the financial and operating network around the company. The round also reconnects Harmony's founders with Lightspeed partner Yoni Cheifetz, who previously worked with them during the Epsagon journey.

The competitive pressure will be real. Incumbent service-management vendors have distribution, embedded workflows, and large customer bases, while newer AI companies can move quickly around specific employee-service tasks. Harmony's answer is to integrate with the systems enterprises already use, establish a context layer across them, and prove that autonomous resolution can become the default rather than an experimental feature.

What This Funding Changes

The $34M gives Harmony room to expand product depth and integrations before the category settles around a small group of winners. Hiring will matter, but integration quality, governance, security, and the reliability of agent actions will matter more because internal services touch sensitive identity, device, application, policy, and financial data.

The company also has to preserve trust as it moves from narrow IT workflows into HR, finance, procurement, legal, cybersecurity, and ERP. Every successful resolution can strengthen organizational knowledge, but every incorrect action carries more weight than a wrong chatbot response. Enterprise buyers will judge Harmony on controlled execution, auditability, and measurable operational outcomes.

For operators, the broader lesson is that AI-native enterprise software is shifting from assisting work to completing bounded processes. Harmony's round is a bet that employee service is one of the best places to make that shift visible because the queues are familiar, the costs are persistent, and the results can be measured. The next phase will show whether Harmony can turn that wedge into the broader enterprise service layer its founders and investors envision.