June Raises $20M to Automate Enterprise AI Deployment

A startup emerging from stealth Monday claims to have cracked a problem vexing Fortune 500 companies: getting artificial intelligence to work inside the tangled infrastructure of large enterprises. June, founded by four engineers with deep roots at Salesforce, raised $20 million in pre-seed funding from Marc Benioff’s Time Ventures, along with backing from Dell Technologies founder Michael Dell, Box CEO Aaron Levie, and CrowdStrike founder George Kurtz.

The company’s core insight is straightforward but high-stakes. Implementing AI agents in corporate environments requires far more than building a model or a chatbot. It demands untangling decades of legacy systems, duplicate data fields, and fragmented workflows — work that currently falls to either armies of expensive consultants or specialized forward-deployed engineers who parachute into companies to stand up AI implementations.

“AI, paradoxically, increases the demand for professional services,” said Efrat Rapoport, June’s founder and a former Salesforce executive. “The industry’s answer to AI implementation is, ‘let’s hire more and more and more people.’” Rapoport and her three co-founders — Ohad Hen, Barak Goldstein, and Idan Tsitiat — believe there’s a better path. Their platform automates much of the discovery, planning, and deployment work that currently requires armies of specialized labor.

A Team Shaped by Inside Experience

The four founders previously built Bonobo AI, a voice-to-text company that launched in 2017. Salesforce acquired the startup in 2019, and the team spent years working on the enterprise’s AI initiatives before spinning out to start June. That insider experience proved valuable to investors. Rapoport said the company “didn’t even have a deck for this raise” — the investors’ confidence in the team was enough.

That background at Salesforce proved instructive. Watching the company’s customers grapple with AI deployment revealed a systematic problem: no amount of model power solves implementation complexity.

The Real Problem: Legacy Chaos

Read also: London’s anti-hustle hacker house challenges Silicon Valley founder culture · The Price of Ambition: How Young AI Founders Handle Relentless Pressure · Ellis AI Raises $10M to Automate Fragmented Private Credit Operations · Meta Ordered to Pay $567M and Restrict Features for New Mexico Minors

Enterprise software environments are rarely clean. Companies running Salesforce, ServiceNow, Databricks, Workday, and a dozen other platforms find their data spread across disconnected systems. “Before AI can create value, someone has to deal with legacy systems,” Rapoport explained. “You have fragmented data across these platforms. You have complex workflows. You have years of technical debt.”

The specific challenge is that AI agents need to navigate this mess reliably. “How does an agent know how to operate when you have 10 duplicate database fields that say the same thing, and different teams are using them?” Rapoport asked. Building a theoretical agent is easy. Making it work when production data is messy, policies are tangled, and nobody fully understands what inherited systems actually do — that’s the hard part.

This gap has given rise to an entire category of specialists. Forward-deployed engineers, often commanding six-figure salaries or equivalent consulting fees, are now in such high demand that their scarcity has become a bottleneck for companies betting on AI.

How June’s Platform Works

Automated Discovery and Planning

June’s platform starts by scanning a company’s existing systems to map business processes, identify bottlenecks, and understand data flows. It then generates a detailed, step-by-step roadmap for implementing AI agents that align with the company’s actual infrastructure — not how the company wishes it were organized.

“We give you the full roadmap automatically of what needs to happen step by step for you to actually implement this agent successfully in an enterprise environment, which is often very complex,” Rapoport said. “We give you a step by step guide. ‘Remove these duplicates. Connect to this data source.’ And then you click on ‘build’ on each task, and June starts building it for you in the organization.”

Execution and Integration

Once the plan is generated, the platform doesn’t just advise — it executes. It builds the connections, cleans the data definitions, and stages the deployment. Teams get clear visibility into each step and can approve or adjust as needed. Integration happens automatically through the company’s existing communication channels, keeping relevant teams informed as work progresses.

Enterprise Safety and Transparency

A key selling point is transparency. Unlike black-box consulting engagements where mysterious experts produce unexplained results, June’s approach documents every decision and lets companies understand exactly what changes are being made and why.

CMG’s Path from Roadblock to Deployment

The value of this approach became clear at CMG, a major U.S. mortgage lender. Paul Akinmade, the company’s chief strategy officer, had moved his software engineering team onto Claude Code and saw immediate promise. But when his team tried integrating Claude Code with Salesforce — a critical system for CMG’s operations — they hit a wall.

The pressure was real. Akinmade had committed publicly at Salesforce’s annual customer conference to returning with 100 agents running across the company’s operations. As his team spent weeks consulting with architects and forward-deployed engineers without making progress, that target looked unrealistic.

June changed the situation. With the platform’s help, Akinmade’s team gained a clear map of where agents could safely operate and what technical groundwork needed to happen first. They could begin deployment confidently, even before formal sign-off from Salesforce.

Replacing Consultants, Not Complementing Them

Rapoport frames June as a tool that works alongside forward-deployed engineers and consultants. But her customers may have different ideas. Akinmade was candid about his preference: “If your product requires FDEs, I don’t want your product. I’ve already done that and I’m getting annoyed by it. I don’t want a black box. I don’t want something only certain people can figure out. I want an easy-to-use tool.”

That sentiment points to a potential market dynamic. As the cost of deploying AI agents in enterprises has become a barrier to adoption, customers are actively seeking solutions that reduce dependency on expensive specialized labor. June’s positioning — delivering clarity, automation, and visibility — directly addresses that demand.

The irony is sharp: AI was supposed to automate professional work and reduce the need for specialized human labor. Instead, deploying AI has created an entirely new category of specialized, expensive professionals. June’s bet is that automating the automation deployment process can break that cycle.