By Sagar Shankaran, Founder of CallSphere
ServiceTitan's Atlas, Dispatch Pro, and Adaptive Capacity make HVAC the most native AI voice agent vertical of 2026. We walk through Job, Appointment, and Technician APIs for a CallSphere build.
Key takeaways
ServiceTitan made AI dispatch native in 2026 with Dispatch Pro, Atlas, and Adaptive Capacity. CallSphere's HVAC deployment integrates the same primitives, so independent voice agents now book directly to the dispatch board without a middleware layer.
flowchart TD
Client[MCP client · Claude Desktop] --> MCP[MCP server]
MCP --> Tool1[Tool: Calendar]
MCP --> Tool2[Tool: CRM]
MCP --> Tool3[Tool: KB search]
Tool1 --> SaaS1[(Calendly)]
Tool2 --> SaaS2[(Salesforce)]
Tool3 --> SaaS3[(Notion)]A homeowner calls at 7:42pm with a no-heat emergency in 18°F weather. The CallSphere voice agent identifies the customer by phone in ServiceTitan, sees prior service history, classifies the call as Emergency Heat, checks Adaptive Capacity for the next on-call technician, books a 9pm window, and texts the customer the technician's name and ETA. Total time: 2 minutes 14 seconds. The dispatcher sees the booked job appear on the board and never has to touch it.
ServiceTitan namespaces resources by domain. CallSphere uses Dispatch (/dispatch/v2/tenant/{tenantId}) for jobs, appointments, and capacity; CRM (/crm/v2/tenant/{tenantId}) for customers and locations; Settings (/settings/v2/tenant/{tenantId}) for technicians; Job Booking (/jpm/v2/tenant/{tenantId}) for the booking flow itself.
Key endpoints: GET /jpm/v2/tenant/{tenantId}/job-types (callable services), POST /jpm/v2/tenant/{tenantId}/jobs (book a job), GET /dispatch/v2/tenant/{tenantId}/capacity (Adaptive Capacity availability), GET /dispatch/v2/tenant/{tenantId}/appointments (current schedule), PUT /dispatch/v2/tenant/{tenantId}/appointments/{id}/assignments (assign a technician).
Hear it before you finish reading
Talk to a live CallSphere AI voice agent in your browser — 60 seconds, no signup.
Auth is OAuth client credentials with a tenant-scoped App Key. Required scopes vary by endpoint.
CallSphere ships an HVAC tool pack mirroring the Healthcare 14-tool pattern: lookup_customer, create_new_customer, get_available_slots, book_job, reschedule_job, get_technician_eta, request_callback, escalate_to_dispatcher, add_job_note, request_invoice, plus emergency_triage, upsell_membership, book_maintenance_visit, and survey_post_job. All 14 wrap ServiceTitan endpoints one-to-one.
CallSphere runs 37 specialist agents and 90+ tools across 6 verticals. Pricing $149 / $499 / $1499. 14-day trial. 22% affiliate program.
/crm/v2/customers?phone={e164}./jpm/v2/jobs with selected window, job type, and customer ID.PUT /assignments only if the dispatcher's policy is to assign at booking time; most teams let Dispatch Pro auto-assign.job.scheduled, job.cancelled, technician.dispatched so the voice agent can confirm or rebook out-of-band changes.// Book an HVAC job in ServiceTitan from a CallSphere voice agent
const job = await fetch(
`https://api.servicetitan.io/jpm/v2/tenant/${tenantId}/jobs`,
{
method: "POST",
headers: {
Authorization: `Bearer ${stAccessToken}`,
"ST-App-Key": stAppKey,
"Content-Type": "application/json",
},
body: JSON.stringify({
customerId: matchedCustomer.id,
locationId: matchedLocation.id,
jobTypeId: 12, // "Emergency No Heat"
priority: "Urgent",
campaignId: 4501, // "Inbound Voice Agent"
summary: "No heat reported. CallSphere voice agent diagnosis: possible ignition issue.",
appointments: [{
start: chosenSlot.startIso,
end: chosenSlot.endIso,
}],
}),
}
);
Does ServiceTitan support third-party voice agents? Yes via the public API. ServiceTitan also offers their own AI Voice Agent in Contact Center Pro, which is fully native; CallSphere is the right pick when you want a multi-vertical voice provider that integrates with ServiceTitan rather than replaces parts of it.
Still reading? Stop comparing — try CallSphere live.
CallSphere ships complete AI voice agents per industry — 14 tools for healthcare, 10 agents for real estate, 4 specialists for salons. See how it actually handles a call before you book a demo.
How does Adaptive Capacity work over the API? GET /dispatch/v2/capacity returns time-windowed booking availability accounting for technician shifts, drive time, and skill matching. CallSphere caches 60s.
What about V1 vs V2 API? Use V2. V1 is in deprecation. The V2 namespace structure (dispatch, crm, settings, etc.) is what current docs cover.
Can the agent reschedule mid-call? Yes via PUT /jpm/v2/jobs/{id} for the date and PUT /dispatch/v2/appointments/{id} for the appointment window.
How do I demo this? Start a trial. Pick HVAC during onboarding or contact us for an HVAC walkthrough.
Written by
Sagar Shankaran· Founder, CallSphere
Sagar Shankaran is the founder of CallSphere, where he builds production AI voice and chat agents deployed across healthcare, hospitality, real estate, and home services. He writes about agentic AI, LLM engineering, and shipping voice agents that handle real calls in production.
See how AI voice agents work for your industry. Live demo available -- no signup required.
Graphiti is the open-source temporal knowledge graph for AI agents in 2026. Learn how bi-temporal memory beats vector RAG for voice agents and long-running LLMs.
How we built a fault-tolerant HVAC emergency triage and tech-dispatch platform on Kubernetes — three-tier CQRS, 11 micro-agents on the OpenAI Agents SDK + LangGraph, NATS JetStream, DTMF/SMS/WebSocket acceptance, circuit breakers, and an evaluation pipeline that catches regressions before they wake a tech at 3 AM.
Self-correction is now a property of the model, not the framework. What that means for production agent reliability, voice/chat fallbacks, and CallSphere.
How to design a multi-agent system using MCP for tools and A2A for cross-vendor coordination, with a CallSphere voice agent as a participating node.
Spring 2026 AC season starts now. With the voice AI market at $47.5B by 2034, HVAC shops without after-hours voice agents will lose to those that have them.
Offline evals catch regressions before deploy on a fixed dataset. Online evals catch real-world drift on live traffic. You need both — here is how we run them.
© 2026 CallSphere LLC. All rights reserved.