GovTech engineering capabilities
Public money, public records and sensitive operations, and how AI has to fit them.
AI in government software only works when the money still balances, the source is always shown and data stays where the agency needs it. These are the systems we have worked with and how the team works.
The team
Who builds it.
Product and workflow analysis
Business analysts who map how payments, records and approvals move through each office before anything changes.
AI engineering
Retrieval with citations, entity extraction, matching across records and self-hosted models, evaluated on your own data.
Platform engineering
Multi-agency tax and payment platforms, gateways and card swipers, month close and distribution, letters and reports.
Releases and operations
Separate QA and production environments per tax office, and tested, versioned releases for environments we cannot access.
Systems
Systems we have worked with.
Built in client products. We scope yours in the Product Review.
| Area | Systems |
|---|---|
| Payments | NMI gateway and webhooks for ACH reversals, card swipers at the counter, tokenized card and bank payments, lockbox and mortgage servicer files |
| Tax and fines | Roll import, change orders, batches, month close and distribution checks, tax sale files, receipts by print, email and text |
| Records and documents | PDF, Word, Excel and scanned records; Surya and PaddleOCR self-hosted; PostgreSQL with pgvector; signed, short-lived viewer links |
| Public data | House Clerk and Senate roll calls, congress.gov, FEC and OpenSecrets data, Census TIGER/Line boundaries |
| Maps and access | Google Maps layers, GeoJSON and KMZ boundaries, JWT sign in, role and permission middleware |
| Cloud and models | AWS (S3, CloudFront, EC2 GPU hosts), DigitalOcean, Docker, vLLM with self-hosted models, on-premise VM releases |
How the team works
One team across product, AI, platform and releases.
A named lead engineer and US-hours overlap, from the first Product Review to production.
Every engagement
- We map how money and records move before we change a live platform.
- We change it in stages, each checked against the flows it feeds.
- Every change goes through your pull request review and our QA.
- We work in your cloud accounts or ship releases for the agency's data center, with access you grant and can revoke.
See it on your own platform.
Bring one product, workflow or AI feature. In 30 minutes we tell you which of the three ways to start fits, what it would cost to serve per agency and what agencies will ask about it.
Calls run in US Central time. No generic AI presentation: we discuss your product, users and production constraints.