Hosted AI Agents, Built and Operated by ITECS
A hosted AI agent serves a whole workflow rather than one person, on frontier models through enterprise API agreements. ITECS builds it, runs it in an environment ITECS operates, and keeps operating it after launch.
- Who evaluates
- COO, CFO, and operations leadership, with an IT director reviewing
- Runs in
- A governed Linux environment that ITECS operates, monitors, and patches
- Connects to
- ERP, CRM, finance, and document systems, using client-managed, least-privilege credentials
- Typical first engagement
- An AI Blueprint, credited toward the build
Deployment patterns
Three ways to run an agent
- Where it runs
- The employee's own Claude, Codex, or Copilot seat. No server.
- Who evaluates it
- Department heads and the employees who will build
- What it connects to
- Files and systems that employee can already reach
- Typical first engagement
- A Guided Build Sprint or a Local Agent Sprint
Hosted AI Agents
- Where it runs
- A governed Linux environment ITECS operates, on frontier models
- Who evaluates it
- COO, CFO, and operations leadership, with IT reviewing
- What it connects to
- ERP, CRM, finance, and document systems through controlled integrations
- Typical first engagement
- An AI Blueprint, credited toward the build
- Where it runs
- A server with data-center class GPUs inside your network, on local models
- Who evaluates it
- CFO, CIO, IT director, and regulated buyers
- What it connects to
- The same systems, with the data staying on premises
- Typical first engagement
- An AI Blueprint and a sizing call for the server
Hosted AI Agents
What a hosted agent is
- Who it serves
- A workflow, not one person. It runs on a schedule or on demand, for the people the workflow involves.
- Models
- Frontier models through enterprise API agreements. ITECS verifies the selected platform's data-use and retention terms, tenancy, and connectors before production.
- Review and audit
- Human review points and an audit trail are designed in. Agents assemble, draft, and flag; people keep judgment and client-facing decisions.
Controls
Seven controls between an agent and your production systems
Each layer answers one question an agent must clear before it acts. The linked ITECS guides go deeper on every control.
- Inventory every agent and dependencyAn AI bill of materials inventories what each agent depends on before it moves to production. AI bill of materials for agents
- Trust only pinned, scoped toolsGoverned registries, pinned versions, scoped access, and drift detection keep poisoned tools, plugins, skills, and MCP servers out. securing the agent tool supply chain
- Authorize sequences, not single actionsAuthorization judges the chain of tool calls an agent makes, not each action alone. AI agent authorization checklist
- Bound every delegation and handoffWhen work moves between agents and people, bounded task contracts, least-privilege access, and evidence-bearing handoffs keep an accountable owner in place. AI agent delegation contracts
- Inject secrets at runtimeScoped runtime injection, for example through the 1Password CLI, reduces how often credentials sit in plain text around agent workflows. using 1Password with custom agent CLIs
- Score tool use before productionTrace-based evaluation scores outcomes, tool use, safety, reliability, latency, cost, and human review before an agent handles live work. AI agent evaluation before production
- Stop, contain, and recoverAn emergency stop, an incident playbook for containment and evidence, and backups an agent cannot delete define what happens when something goes wrong. AI kill switch plan, AI agent incident playbook, AI agent recovery plan
After launch
Operated, not handed off
ITECS runs production agents under Agent Operations, billed under Managed Intelligence Services and never absorbed into the build fee.
- Monitoring and evaluations
- Agent runs, tool calls, costs, and answer quality are monitored and re-scored against the evaluation set.
- Quality and drift checks
- Changes in data, prompts, or model behavior are caught before they erode results.
- Data refresh and exception triage
- Retrieval sources stay current, and exceptions route to the right person with context.
- Production tuning based on usage
- Prompts, tools, and thresholds are tuned from real usage, with changes versioned and tested.
- When the model or platform changes
- Platform and model version changes fall outside the 30-day defect warranty and route to Agent Operations.
See the Managed Intelligence Provider model and the Agent Operations prices for two or more agents.
Pricing
The build ladder
The AI Blueprint sets the actual build rung and its fixed-fee not-to-exceed. The rows below show the typical path and the catalog range for each step.
- FirstAI BlueprintAgent Discovery & Technical Specification. Credited toward the build.$4,500–$7,500
- Prove itProof of Concept / PrototypeA bounded prototype that proves technical feasibility and workflow fit before a production commitment.$8,000–$18,000
- One workflowSingle-Workflow Production AgentA production agent for one primary workflow, deployed with documentation, human-in-the-loop review, and staff handoff.$18,000–$35,000
- Line of businessIntegrated / Line-of-Business AgentA line-of-business production agent for multi-source ingestion, reconciliation, audit trails, validation, exception handling, and reviewer workflows.$35,000–$75,000
- Multi-agentMulti-Agent System / Process RedesignAn orchestrated multi-agent system or end-to-end redesign of a major workflow with integrations, monitoring, and rollout enablement.$55,000–$120,000
- After launchAgent Operations, one production agent$2,500/mo
Prices from the published AI pricing catalog.
Example agents
Designs that suit Hosted AI Agents
Each is a starting design. An AI Blueprint adapts it to your goals and the systems it will connect to.
- PPV Agent
Explains whether you are paying what you expected, why purchase price variance moved, and what exposure comes next.
- Manufacturing
- Finance
- Hosted AI Agents
- Private AI Servers
- Field Examination Analyzer
Produces the standard working-capital exam and a first-draft report from borrower data, so examiners move straight to findings.
- Financial Services
- Finance
- Hosted AI Agents
- Private AI Servers
- Portfolio Monitoring & Covenant
Turns periodic borrower reporting into continuous covenant and collateral monitoring, with every alert traceable to its source.
- Financial Services
- Finance
- Hosted AI Agents
- Private AI Servers
- Pre-Bill Review Agent
Checks proforma time entries against each client's billing guidelines and rate letters, and drafts fixes for the timekeeper and billing partner to approve before release.
- Legal
- Finance
- Hosted AI Agents
- Private AI Servers
- Tax Return Readiness Agent
Checks each client's uploads against last year's return and the organizer, tracks expected K-1s, and drafts the missing-items request for staff approval before a preparer starts.
- Accounting
- Front office
- Hosted AI Agents
- Private AI Servers
- WIP & Change-Order Agent
Reconciles job cost, commitments, and pay applications into a traced monthly WIP, and finds performed change work that was never billed, for the PM, controller, and CFO to approve.
- Construction
- Finance
- Hosted AI Agents
- Private AI Servers
- Prior Authorization & Denial Agent
Verifies coverage before the visit, checks prior-authorization requirements, assembles cited payer packets, and ranks and drafts denial appeals, for staff and clinicians to approve.
- Healthcare
- Operations
- Hosted AI Agents
- Private AI Servers
- Joint-Interest Cost Agent
Checks field tickets and vendor invoices against MSA rates, AFEs, the JOA, and the field record before the JIB goes out, and routes each exception with its evidence to the JIB accountant.
- Oil and Gas
- Finance
- Hosted AI Agents
- Private AI Servers
- Sales Pipeline Agent
Researches leads, drafts outreach, logs approved activity, and supports scoring inside your CRM.
- All industries
- Sales
- Hosted AI Agents
- Back-Office Workflow Agents
Handles lead follow-up, invoice intake, and data entry inside the tools your team already uses.
- All industries
- Operations
- Local AI Agents
- Hosted AI Agents
FAQ
Hosted AI Agents FAQ
Where does our data go?
Data moves between your systems and the ITECS-operated environment through the integrations the AI Blueprint specifies, and to the model provider under enterprise API terms. ITECS verifies the selected platform's contractual data-use and retention terms, tenancy, connectors, and required data movement before production, and non-contracted consumer services are excluded.
Which model does the agent use?
ITECS is platform-agnostic and recommends the model that fits the workflow, the data, and the governance requirements. The AI Blueprint records the choice and the reason for it.
Who holds the keys?
Client systems use client-managed, least-privilege credentials. Keys and access tokens are kept in encrypted vaults and injected at runtime, not hardcoded or shared between client environments.
What happens when the model changes?
Platform and model version changes fall outside the 30-day defect warranty and route to Agent Operations, which covers monitoring, evaluations, quality and drift checks, data refresh, exception triage, and tuning.
How is a hosted agent priced?
Every build starts with an AI Blueprint, credited toward the build. The build is then quoted on one rung of the published ladder with a fixed-fee not-to-exceed. Agent Operations after launch is billed separately under Managed Intelligence Services.
Start with an AI Blueprint
The AI Blueprint reviews your workflow, data sources and integrations, specifies the architecture and the human-review model, and ends with a fixed-fee not-to-exceed for the build.