Services

Three engagements, one focus: AI connected to your systems.

Most clients start with a readiness assessment, move to a first integration, then keep Telepath on to run and extend it. You can also come in at any step.

01 · Start here

AI Readiness & Governance

A fixed-scope, fixed-price assessment. Before anything gets built, we work out where AI will pay off and what your leadership, IT and compliance teams need to see before they approve it.

Use-case inventory, ranked by value and effort
Data classification: what can go to a frontier model, and what stays in-house
AI risk assessment aligned to NIST CSF 2.0
Acceptable-use policy, change controls and vendor review
A scoped build plan for the first integration
02 · Build

AI Integration

An assistant, portal or agent that works on your real data, inside your own cloud tenant, under your own identity and access rules.

Connected to the systems you run (see the list below)
Microsoft Entra sign-in, role-based access and Conditional Access
Document, deck and spreadsheet generation in an isolated sandbox
Usage, cost and performance telemetry from day one
Staging and production environments with repeatable deploys
03 · Run & improve

AI Operations & Leadership

AI systems aren't finished at launch. Models improve, users ask for more, and costs need watching. This retainer keeps the platform moving forward.

Model upgrades, tested against your own benchmarks
New integrations and features, released on a steady cadence
User feedback routed into your service desk and turned into release notes
Monthly usage and cost reporting for leadership
Fractional AI leadership: board updates, vendor evaluation, team coaching
Integrations

What we connect AI to.

An assistant is only as useful as what it can reach. These are the systems we’ve already wired into production AI.

Microsoft 365 & Entra ID

Single sign-on, groups and app roles, the Graph API, SharePoint and Outlook.

Data warehouse & SQL

Azure SQL and operational data stores, with read-only, row-scoped access for the model.

Service desk & ITSM

Ticket search, trends and ticket creation through the service desk API, with links that go straight to the ticket.

Network & security tools

Firewall and device-management telemetry turned into plain-language status and alerts.

CRM & line-of-business apps

Synthetic monitoring of how those apps actually perform for users, with results the assistant can explain and report on.

Documents & files

Read what users attach, and produce finished Word, PDF, PowerPoint and Excel files in your brand.

Security & Governance

Governance built in, not bolted on.

Our governance practice was proven in SEC-regulated financial services, under GLBA and FFIEC expectations. Every client gets the same controls, whether you answer to regulators, a board, or simply your own standards.

Talk Through Your Requirements
The platform, its data and its logs run in your own cloud tenant
Zero data retention with the model provider, so prompts and responses aren’t kept after each reply
Data sources gated by role, so each user reaches only what their role allows
Generated code executes in a disposable sandbox with no network access
Every turn logged: who asked, which model, what it cost, how long it took
Written security architecture your IT team, auditors or regulators can review

Not sure where to start? Tell us which process eats the most hours, and we’ll tell you honestly whether AI can help.

Get in Touch

Also available to integration clients: ongoing managed IT support, including monitoring, Microsoft 365 administration and helpdesk, for the systems underneath.