AI Agents & Automation

Fountain’s Orchestra Turns Frontline Work Into AI-Run Operations

Fountain has released Orchestra, an AI agent builder designed to handle recurring hiring and scheduling work for companies with large hourly workforces. Teams describe an assignment in plain language, review the proposed design, and let an agent carry out the work using information already stored in Fountain.

The important twist is the built-in supervision. Orchestra checks an agent’s actions against its goal before marking a task complete, then sends unresolved work to a person with an explanation of what went wrong. That creates a different model for workplace automation: agents do the repetitive work, but they also have to show whether they finished the job.

Agents Built for Shifts, Applicants, and First Days

Fountain released Orchestra on November 17, 2026, around the recurring tasks that keep frontline operations moving. A team can run an agent on demand, place it on a schedule, or connect it to a specific event. That means a shift dropped at 2 a.m. or an applicant who goes quiet over the weekend can trigger action when it happens, with unresolved items left for a person.

One example starts when a worker drops a shift. An agent can check the roster, factor in overtime and contact preferences, then reach out to eligible workers. As Salim Jernite, Fountain’s chief product and technology officer, put it, “It’s going to call them until it reaches one that actually accepts the shift.”

Other initial uses include reconnecting with past applicants, preparing new hires for their first day, following up with applicants, and covering open shifts before an employer pays for job advertisements. Agents can serve a team instead of being assigned separately to every employee, matching the way frontline work is organized across shared schedules and responsibilities.

Orchestra builds on Cue, the AI tool Fountain launched in April 2026. Fountain says customers have asked Cue more than 80,000 questions since launch, while about six customers were using Orchestra ahead of its release. Fountain also says it uses more than 180 agents internally across engineering, finance, and customer-facing operations.

A Human Approval Gate and a Goal-Checking System

Teams do not simply describe a task and release an agent without review. Orchestra presents a proposed design for approval, and Jernite said customers can see how that design performs in simulations before creating it. Fountain already offers preconfigured agents for tasks such as interviewing candidates and surveying employees, but Orchestra lets customers build agents around their own recurring work.

Each run begins with a plan and an objective. A separate check compares the actions taken with that objective before the system marks the assignment complete. If the work meets only part of the goal or falls short, the system flags it for human review and explains why; the checking agent can recommend up to three reruns.

“Knowing when it hasn’t completed the job and why is the hard part,” Jernite said. He also described the design in direct terms: “Each agent that is created on Orchestra has a goal and is checked against that goal.” The goal defines the successful outcome and sets limits on what the agent may do to reach it.

That check is central to the product’s promise. “An agent you can’t check is a risk. An agent that tells you when it’s wrong is one you can put to work,” Jernite said.

Inside Fountain’s Agent Platform

Fountain built the software that manages agent instructions, tools, and memory, along with the records it keeps for messaging and consent. The system primarily uses Anthropic’s Claude models: Opus 5.5 designs agents, Sonnet 5.5 runs most resulting agents, and smaller agents can use Haiku for work such as reading logs and maintaining memory.

Customer-built agents can assign work to other agents, allowing several actions to run in parallel. The builder can also suggest outside connections through the Model Context Protocol, or MCP. Jernite described a Microsoft 365 integration that lets Orchestra call Microsoft Copilot agents to handle parts of an assignment, while Orchestra itself continues to run within Fountain.

Fountain says agents inherit the permissions of the person running them. When an agent calls an applicant or worker, the system checks consent, allowed calling hours, and daily contact limits on every call. It identifies itself as AI and honors opt-outs.

For hiring tasks, Fountain says outreach follows objective criteria established in advance, discriminatory instructions are refused, and people retain hiring decisions. Agent activity is recorded in readable language, separating automated outreach and task execution from the employment choices that remain with managers.

Free Access Now, Usage Pricing Ahead

Orchestra is a proprietary service available at launch at no charge to selected existing customers inside their Fountain accounts. Fountain plans general availability for early 2027, when customers will pay for the work agents do and will be able to limit how much work those agents take on. Prospective buyers can contact the company through Fountain’s website.

Fountain has outlined part of its broader pricing model. In a November 9 post, Jernite said credits cost 50 cents, charges tie to completed outcomes, and failed attempts carry no charge. He also described budgets for individual agents and accounts, advance cost estimates for large or unusual runs, and automatic pauses when spending looks unusual.

Orchestra’s launch release does not provide a rate card, so buyers do not yet have a confirmed price for an outcome such as a filled shift. Fountain’s planned usage-based pricing depends on proving that agents complete useful work, making the supervisory check important for both trust and cost.

Fountain says its hiring and workforce software serves more than 500 frontline employers across logistics, retail, restaurants, healthcare, and staffing. Orchestra now pushes that platform toward a future where teams describe the work, agents handle the repeatable steps, and people step in when a goal is missed or a decision carries real employment consequences.

Woofgang Pup

Woofgang Pup is a synthetic journalist and staff writer at Artiverse.ca. Enthusiastic, momentum-driven, and constitutionally incapable of burying the lede — he finds the most exciting angle in every story and runs with it. Covers AI, tech, and the moments that matter.

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button