Skip to main content
FirstCoast.ai
All case studies
Case study: Professional services agency

An Agentic Operating System automating and overseeing all aspects of a Professional Services Agency

A services company rebuilt its back office as a fleet of AI agents under one operating system: a version-controlled knowledge base as the shared brain, an approval surface in Telegram on the owner's phone, and an Orchestrator agent that adds or retires specialists as each job demands. Nothing reaches a client, or goes out publicly, without a human tap.

How the agentic operating system is put togetherOne version-controlled git repository holds the company's services, pricing, voice, procedures, and client context, and every agent reads from it, so an answer drafted anywhere matches what the company offers and charges. An Orchestrator agent reads each new job and decides which specialists it needs, so the working fleet grows or shrinks with the work instead of running a fixed roster, across five functions. Marketing runs weekly LinkedIn content drafts pulled from the week's real work. Sales handles lead intake and scoring on every new lead from HubSpot, and produces follow-up drafts. Billing runs invoice drafting from QuickBooks Online and retainer billing on schedule, and produces invoice drafts. Operations runs two agents: inbox operations over the day's mail in Gmail, which produces reply drafts, and meeting notes, which turns every Zoom call into a record. Reporting runs the morning brief on every active project before the workday starts, and produces a project status. The four outbound streams, LinkedIn drafts, follow-ups, invoices, and replies, all converge on one approval queue, where every social post, email, invoice, and proposal waits as a draft until a human reviews and releases it. Only released work reaches a client system or goes out publicly: HubSpot, QuickBooks Online, Gmail, and the company's own channels, the same systems the agents read from. Nothing has ever gone out unapproved, because the gate is architecture, not policy. The other two streams never leave the building and never touch the gate: the meeting record, with its action items tagged by owner and confidence level, and the morning brief. Meeting records are filed back to the git repository, which is the system of record, closing the loop on the canon. The whole fleet still costs a few dollars a month in metered AI spend, because every workflow is costed per run before it is built, many use no AI at all, and the rest use the smallest model that does the job: fast models for classification, stronger ones for drafting.THE SHARED BRAINGIT REPOOne version-controlled knowledge base, read by every agentServicesPricingVoiceProceduresClient contextso an answer drafted anywhere matches what the company offers and chargesAN ORCHESTRATOR SIZES THE FLEET TO THE WORKMARKETINGWeekly LinkedIn draftsfrom the week’s real workLinkedIn draftsSALESLead intake and scoringevery new lead, from HubSpotFollow-up draftsBILLINGInvoice draftingfrom QuickBooks OnlineRetainer billingruns on scheduleInvoice draftsOPERATIONSInbox operationsthe day’s mail, in GmailMeeting notesevery call, from ZoomReply draftsMeeting recordREPORTINGMorning briefevery active projectbefore the workday startsProject statusEVERY OUTBOUND DRAFTTELEGRAMDRAFTS ONLY, NEVER AUTOMATICOne approval queueEmail draftsInvoice draftsProposal draftsLinkedIn draftsa human reviews and releases every one before it goes anywhereNEVER FACES A CLIENTMeeting recordaction items, every callOwnerConfidenceMorning briefevery active projectbefore the workday startsTHE GATE IS ARCHITECTURE, NOT POLICYClient systems updatedZERO UNAPPROVED, EVERthe same systems the agents read fromHubSpotQuickBooks OnlineGmailLinkedInTOTAL METERED AI SPENDA few dollars a monthevery workflow costed per run before it is builtNo AI at allFast: classificationStronger: draftingFILED TO THE RECORD

The challenge

  • The owner was the operating system: every lead, invoice, follow-up, and status check ran through one person's memory and working hours
  • Admin work like billing, triage, meeting notes, project status, and social content consumed hours that should have gone to clients
  • Off-the-shelf tools each solved one task but shared no context, so nothing knew what the rest of the business knew

What we built

An agentic operating system. Company knowledge lives in one version-controlled repository every agent reads from: services, pricing, voice, procedures, and client context. An Orchestrator agent reads each new job against that knowledge base and decides which specialists it needs, so the fleet runs the day-to-day: lead intake and scoring, invoice drafting, retainer billing, Zoom meeting notes, inbox operations, weekly LinkedIn content drafts, and a morning brief that reports the state of every active project before the workday starts.

  • One knowledge base grounds every agent, so an answer drafted anywhere matches what the company actually offers and charges
  • An Orchestrator agent decides which specialists a job needs, so the working fleet grows or shrinks with the work instead of running a fixed roster
  • A Zoom transcript becomes a meeting record with tracked action items, each tagged with an owner and a confidence level, so promises made on a call stop living in memory
  • A marketing agent drafts LinkedIn posts from the week's real work, so the social calendar costs editing minutes instead of writing hours
  • Every outward-facing email, invoice, proposal, and social post is created as a draft and waits for human approval, never sent or posted automatically
  • Each agent uses the smallest AI model that does the job, and many workflows use no AI at all

How it's built

A git repository as the system of record, an automation layer for scheduled and event-driven workflows, Telegram approvals, a Postgres database, and integrations with HubSpot, QuickBooks Online, Zoom, Gmail, and LinkedIn content drafting. An Orchestrator agent reads each new job against the knowledge base and decides which specialists to run, so the working fleet is sized to the task rather than fixed. Model choice is per task: fast models for classification, stronger ones for drafting. Every workflow gets a per-run cost estimate before it is built.

Results

  • An Orchestrator agent sizes the working fleet to the job, running specialists across sales, marketing, billing, operations, and reporting, plus two chat agents serving real users around the clock
  • Total metered AI spend: a few dollars a month
  • Zero automated messages have ever reached a client, or gone out publicly, without human approval, because the gate is architecture, not policy

What would your business look like with an operating system?

The same architecture scales down to a single workflow. The Discovery Audit finds the right first agent.