AI Agents Move From Chatbots to Workflow Owners
Enterprise AI agents are software systems that live inside tools workers already use, understand organizational context, and autonomously complete HR, IT, finance and customer workflows instead of only answering questions. That shift—from static chatbot responses to end‑to‑end task execution—defines the next phase of employee support automation and customer engagement in large organizations. Harmony and Encore AI are early proof points that this is more than hype; investors are paying real money for AI agents enterprise platforms that aim to own workflows, not chat windows. The headline is the funding: Harmony raised a USD 34 million (approx. RM156.4 million) seed round, while Encore AI closed USD 30 million (approx. RM138.0 million) in Series A workflow automation funding to push autonomous agents deeper into back‑office and customer‑facing processes.
Harmony: AI Agents Where Employees Already Work
Harmony’s bet is simple and smart: if you want to fix employee support automation, you go where employees already type, not where IT wishes they would go. Founded by Nitzan Shapira and Ran Ribenzaft—who previously sold Epsagon to Cisco for roughly USD 500 million (approx. RM2.3 billion)—Harmony is a Slack AI integration and Microsoft Teams AI agent platform built to handle IT tickets, HR questions, procurement requests, finance queries, legal approvals, onboarding, app access and password resets. It emerged from stealth on July 28 with a USD 34 million (approx. RM156.4 million) seed round led by Lightspeed, joined by several other investors. The strategic insight is to bypass separate portals and ITSM front‑ends: “You don't ask employees to open another portal. You meet them where they're already typing.” That is exactly where Slack and Teams have become the operating system of knowledge work.
The technology story matters even more than the Slack and Teams logos. Harmony’s core is an organizational context graph that encodes who a user is, their role, permissions, device and work history. Without that, you have yet another chatbot; with it, Harmony argues, you have a system that can resolve requests—reset a password, approve an expense, grant app access—instead of handing the employee a link. Early marketing claims include more than 100 prebuilt agents, deployment in days, and an AI ITSM example citing 48% no‑touch resolution by week two and 75% ticket deflection by month three, though those figures are self‑reported and unverified. The money will go toward engineering and sales hiring, which is necessary because touching IT, HR, finance, procurement, DevOps, security and legal at once is both the opportunity and the risk.
Encore AI: Turning Customer Conversations Into Revenue
If Harmony is about internal tickets, Encore AI is about external money. Encore AI, formerly Insait IO, announced on July 29 a USD 30 million (approx. RM138.0 million) Series A led by Team8, Planven and The Garage to grow an enterprise AI platform that generates revenue from customer interactions rather than deflecting them. Where most contact‑center automation obsessively deflects calls, Encore AI leans into sales and completion: its agents convert leads, close applications and drive upsell opportunities across voice, chat, IVR and live form‑fill channels. The platform’s patented Interaction Mining studies an organization’s top performers, extracts which actions drive results and deploys them as AI agents that operate autonomously or alongside live teams, across channels and languages. That framing matters for banks, insurers and healthcare providers, which Encore AI targets with a compliance‑ready architecture to satisfy strict regulatory expectations.
The funding story doubles as a go‑to‑market signal: Encore AI said the capital will be used to expand global deployment of its platform, and several large commercial banks and insurers moved from customer to investor. The company claims go‑live in weeks, ingesting calls, chats, emails and CRM data without months of configuration. This is the other side of AI agents enterprise adoption: while Harmony makes it easier for employees to get IT and HR help, Encore AI turns years of interaction logs into a playbook that runs on autopilot at scale. Agentic AI is already shifting from pilots to production across customer channels, delivering measurable gains in resolution speed, cost efficiency and sales workflow automation, with AI agents resolving up to 40% of inquiries across chat, email, voice and WhatsApp according to reported figures.
Why Investors Are Backing Agentic Platforms Now
USD 64 million (approx. RM294.4 million) of fresh capital into two AI agent platforms in two days is not an accident; it is a statement that workflow‑aware agents are the next enterprise battleground. Harmony is entering a market where established players already own large pieces of enterprise service management, and where committing to a platform creates deep integrations and high switching costs. Yet those incumbents often require employees to live inside separate portals. Harmony’s pitch is that the workflow should come to the employee inside Slack and Teams, reducing adoption friction and making automation feel like part of everyday chat rather than a new system. Encore AI reflects the same philosophy for customer channels: instead of generic bots, its Interaction Mining tries to encode how top agents sell and comply, then scale those patterns. In both cases, investors are betting that context plus autonomy beats generic conversational AI.
The timing is helped by a broader shift: agentic AI is moving from pilot projects to production and is associated with measurable improvements in resolution speed, operational cost and sales automation. At the same time, Slack and Teams have quietly become the front door for many enterprise AI products at once, from embedded copilots to third‑party AI coworkers, making them ideal real estate for employee support automation tools. Harmony and Encore AI are not early‑stage proofs of concept; they arrive with credible founders, concrete product claims and a clear focus on automation that closes loops. Their challenge will be differentiation in crowded categories and executing without creating new security or compliance problems, especially as large platform vendors push their own agent strategies.
The Future of Enterprise Workflows: Agents in the Flow, Not on the Side
The deeper story behind Harmony and Encore AI is that the interface for work is shifting from forms and portals to conversations where AI agents can quietly complete tasks. Harmony’s organizational context graph is an attempt to give Slack AI integration enough knowledge and permissions to close IT and HR loops without human intervention. Encore AI’s Interaction Mining shows the same pattern in customer‑facing environments: study what works, turn it into autonomous behavior, and run it at scale. This is a meaningful break from the era of static knowledge‑base chatbots. Without context and authority, those tools could answer but not act. With workflow‑aware agents embedded in communication platforms, action becomes the default. Enterprises that treat these systems as core infrastructure for employee support and customer journeys—not as sidecar pilots—will be the ones that see real returns. Everyone else risks owning a very smart chatbot that still leaves humans holding the ticket queue.
