Harmony, an AI platform delivering personalized employee support across IT, HR, finance, and more, announced $34M in seed funding led by Lightspeed Venture Partners. The funding will accelerate product development and expand Harmony toward a broader Enterprise Service Management platform.
SynthePulse Insight · AI deep reading
Version 1 · 1 source
Harmony announced a $34 million seed round led by Lightspeed, with AI agents embedded in Slack and Teams claiming to deflect over 75% of support requests within three months. However, these figures come from a company press release, lack independent verification, and the definition of 'deflection rate' and long-term effectiveness remain unclear.
Harmony is an AI platform providing personalized employee support across IT, HR, finance, procurement, and legal. On August 4, 2026, the company announced a $34 million seed round led by Lightspeed Venture Partners, with participation from Hitachi Ventures, Fin Capital, Mercer Ventures, and Operator Partners, and angel investors including the founders of Wiz and the CEO of Eon.io.
Harmony was founded by Nitzan Shapira and Ran Ribenzaft, who previously co-founded the cloud observability company Epsagon, which was acquired by Cisco for $500 million. Harmony's AI agents are deployed directly in Slack and Microsoft Teams, utilizing a context graph that connects employee identity, devices, applications, and work history to execute actions across enterprise systems.
The company claims to have over 100 pre-built agents that can be deployed in days, not months. Its goal is to shift employee support from a 'ticket and queue' model to 'instant resolution by AI agents,' and ultimately become a comprehensive enterprise service management platform.
Harmony's press release cites customer examples including n8n, eToro, Cyera, and KITH, claiming these customers deflected 48% of support requests within two weeks of deployment and over 75% within three months. Additionally, deflection rates for HR, procurement, DevOps, security, and finance are as high as 68%, 58%, 47%, 42%, and 36%, respectively.
The company also mentions that some deployments have employee adoption rates exceeding 80%, and claims an overall 'no-touch resolution rate' of 70%. These figures come from the company's press release, are source claims, and are not independently verified, with the specific definitions of 'deflection rate' and 'no-touch resolution rate' undisclosed.
The press release also notes that employee support requests typically take up to 48 hours to resolve, and Harmony aims to reduce resolution time from days to minutes. However, this baseline data also comes from the company and lacks third-party corroboration.
Harmony stated that the new funds will be used to accelerate product development, expand the team, and deepen integrations with enterprise software ecosystems. The company plans to expand AI capabilities to cover infrastructure and operations, enterprise applications, cybersecurity, and enterprise resource planning.
Lightspeed partner Yoni Cheifetz said in the press release that Harmony is driving the shift of employee support from 'tickets and queues' to 'instant resolution by AI agents,' and believes the company has the potential to become 'one of the defining enterprise software platforms of the next decade.'
This assessment is the investor's perspective, not independent analysis. In terms of market context, the press release mentions that employee expectations are bringing the personal use experience of ChatGPT and Claude into work tools, but no specific market data is provided.
All key data (deflection rates, adoption rates, no-touch resolution rates) come from the company's press release and are source claims, not confirmed facts. The press release does not disclose the calculation method for 'deflection rate,' sample size, customer industry distribution, nor does it provide independent audits or third-party verification.
Statements like 'deployment in days' and 'resolution time reduced from days to minutes' lack specific case details and may vary by customer environment. Additionally, the press release does not mention any failures or limitations, suggesting possible selective disclosure.
Harmony's claims of 'over 100 pre-built agents' and '70% no-touch resolution rate' are self-descriptions without technical details or benchmark tests. Therefore, these numbers should be viewed as marketing claims rather than verified performance metrics.
This report is primarily based on Harmony's press release, and all performance data (such as deflection rates and adoption rates) are self-reported by the company and not independently verified. Investor assessments and founder backgrounds are public information, but the figure of 'Epsagon acquired by Cisco for $500 million' comes from the press release and has not been third-party confirmed. Readers should treat these data with caution and look forward to independent verification.
Harmony's funding and product positioning reflect the trend of AI agents in enterprise services, but its claimed high deflection rates and rapid deployment lack independent verification. When evaluating such products, one should demand auditable metric definitions and third-party test results rather than relying solely on press release data.
Primary report
Primary source