Permit, intake and casework automation
AI agents for government intake, casework and permits, with staff making every decision
Agents for intake, case notes, permits and correspondence that do the routine steps, follow your approval rules and keep a full audit trail. Staff make the decisions.
Free · about one week · view-only · no obligation
View-only access your IT administrator approves and removes. Nothing in your systems changes.
On this page
Short answer
We build AI agents for public agencies on Microsoft 365, Power Platform and Azure that take on high-volume mandated work: intake, case notes, permit checks and correspondence. Every agent has a named owner, approval rules, logging and an off switch, and staff make the decisions. A pilot agent takes 6 to 8 weeks and typically costs $25K to $50K.
Starting points
Where AI agents for government earn their keep
The cost of waiting
Agents are arriving faster than the controls around them
An agent without an owner, a log and an off switch is a liability.
Ours ship with all three, and they act inside your approval rules.
What it really is
An agent is a new member of staff. It needs a role, a supervisor and a record of what it did.
The software is the smaller part. The work is defining what the agent may do and proving it did only that.
Your options
Scripted automation, an agent platform, or a governed build
Agencies have automated routine work for years. Agents handle what scripts could not, and need more control.
| Scripted automation | Agent platform license | Governed agent in your own Microsoft 365 (us) | |
|---|---|---|---|
| Unstructured documents | No | Yes | Yes |
| Approval rules | Scripted | Configurable | Your rules, with people approving decisions |
| Audit trail | Logs | Varies | Every step logged in your own Microsoft 365 |
| Owner and off switch | Not built in | Varies | Named owner, off switch, review schedule |
| Runs on | A separate platform | The vendor’s platform | Power Platform and Azure you already license |
What is included
Where agents help
Intake
Read applications and submissions, check they are complete, and open the case with the right data attached.
Case notes and summaries
Draft case notes and summaries from meetings, calls and documents for staff to review and approve.
Permits and licensing
Check applications against requirements, request missing items and prepare a decision for review.
Correspondence
Draft letters and emails in the agency’s voice, routed for approval before anything is sent.
Governed and supervised
Each agent has an owner, approval rules, logs and an off switch, and is reviewed on a set schedule.
Measured
Time saved, backlog and error rates reported for every agent, so you know what it is worth.
Our approach
One process at a time, proven before it goes live
Agents earn trust in shadow mode before they handle real work.
- Pick one processChoose the backlog with the clearest rules and the most volume.
- BuildThe agent, its approval rules and its audit trail, in your own Microsoft 365.
- Shadow modeThe agent drafts while staff still do the work, and we compare results.
- Go live and measureSwitch over when accuracy meets the agreed bar, and report monthly.
Our promise: no agent makes a decision that affects a resident. It prepares the work, and a person decides.
Governance, security and data handling
Where your data goes, and who stays in control
- Inside your own Microsoft 365. We build in your Microsoft 365 and Azure environment, using the identity and permissions you already govern.
- Exposure fixed first. An AI assistant sees whatever its user can see, so oversharing and unlabeled sensitive content are dealt with before anything is switched on.
- Hosted in Microsoft Foundry on Azure. We use the model that fits the job: OpenAI, Anthropic or open-source models, deployed in Microsoft Foundry on Azure. Your data is never used to train them (see Microsoft’s data terms for Foundry). Anthropic operates its Claude models on Azure under its own commercial terms, and we show you which model and which terms apply before work starts.
- People decide. Approval rules, audit trails and human review wherever an output affects a resident, a record or a decision.
What it runs on
Built on AI-powered managed services
Agents need running after go-live: monitoring, model updates, access reviews and reports. Our managed service does that alongside your security and licenses.
Proof
Government work this builds on
Agents run on the same workflow platform as the systems we already build for agencies.
The offer
One week, read-only, and a plan you keep
The AI readiness assessment shows what an AI assistant could find in your Microsoft 365 today, which uses are ready, and what each one needs before it goes live. The report is yours whether or not you build with us.
Exposure map
What an assistant could reach today: oversharing, permission sprawl and unlabeled sensitive data, ranked by risk.
Use-case shortlist
Three to five uses your agency could start with, scored for value, risk and readiness.
Plan and fixed price
The controls each use needs and a fixed-price estimate for the first build, ready for your procurement process.
- Free, with no obligation
- About one week
- View-only: nothing in your systems changes
- The report is yours to keep
Why free? It is how we scope a fixed price we can stand behind, and it shows you what to fix whether or not you hire us.
How it runs
Day 030-minute scoping call with the architect who would run the work.
Days 1–2Your administrator grants read-only consent. About an hour of your team’s time.
Days 2–4We review your Microsoft 365 and talk to the use-case owners you choose.
Day 5Findings session: exposure map, shortlist and fixed-price plan.
We reply within one business day.
Read-only Microsoft Graph permissions only, consented by your administrator and revoked at the end. No content leaves your Microsoft 365. NDA and data processing terms before any access.
Why agencies choose us
A Microsoft partner that builds and runs it

Transparent investment
The assessment is free. Here is what the work costs.
Every agency is different, and the assessment gives you the firm number. You deserve a real range first.
| Engagement | Typical timeline | Typical investment |
|---|---|---|
| Pilot agent: one process, shadow mode | 6–8 weeks | $25K–$50K |
| Each further production agent | 1–3 months | $20K–$100K |
| Multi-process deployment | 3–6 months | $75K–$250K |
| Managed AI operations | Monthly | $2.7K–$10.5K a month |
Want a figure for your organization? Get an estimate in 2 minutes with our pricing calculator.
Model usage, Power Platform licensing and training are in the estimate, not a change order.
Common questions
AI permit review and case management AI: questions agencies ask
Can AI review a document or application?
Yes, for the routine checks: an agent can confirm that every required field and attachment is present, compare stated values with the rules for that application type and list what is missing, citing the page. A member of staff reviews its findings and makes the decision.
Can AI read construction plans?
It can read plan sheets and extract what the intake checklist needs, such as sheet index, stamps, dimensions and notes, and flag missing sheets. AI permit review speeds up completeness checks; code compliance decisions stay with your plans examiners, who see the agent’s findings next to each sheet before they decide.
How can AI help social workers?
Mostly by giving back time spent writing: case management AI drafts case notes from the record, prepares letters and summarizes a file before a visit, working from the case management system you already use, and the caseworker edits and approves every word. Decisions about people stay with people.
What is agentic AI in government?
Agentic AI in government means AI agents that carry out steps of a process, such as checking an application, drafting a letter or updating a case, rather than only answering questions. AI agents for government need what any new member of staff needs: a role, a supervisor who approves their work and a record of what they did. Every agent we build has all three, plus an off switch.
Questions buyers ask
Before you decide
Will agents replace staff?
They take the routine steps off staff, such as chasing missing documents or drafting first versions. Decisions stay with people, and most agencies use the time to clear backlogs.
What if an agent gets it wrong?
It runs in shadow mode first, people approve its outputs, its error rate is reported, and it can be switched off in one step.
Will it work with our case management system?
If your system has an API or a Power Platform connector, usually yes. The assessment confirms it before you commit.
Who is accountable for an agent?
A named owner in your agency, supported by us. The owner approves its scope, sees its reports and decides when it changes.
Clear the backlog, keep the control
The free assessment finds the process where an agent would save the most time and prices a pilot.

