FireOS, compliance operations · 2026
Fire-safety compliance was running on spreadsheets, phone calls and memory
Built by Vyzma. Led by Vandan.

Situation
Fire and life-safety agencies run statutory work (Form A, Form B, S.45A audits) with deadlines that slip and defects that go untracked, and revenue leaks with them.
What we built
A command centre: an offline-first field app, a rules engine that blocks non-compliant filings at submission, and a workflow that turns every defect into a work order.
What changed
Statutory cycles are tracked as a system rather than remembered, and a filing that fails the rules cannot be submitted.
The problem
Agencies that inspect and certify buildings carry a statutory calendar for every site. Each cycle has forms, audits and defect lists, and the record of all of it lived in spreadsheets and people's heads.
A missed cycle is a compliance failure for the client and lost revenue for the agency. A defect found on site but never raised as work is both.
What we built
- An offline-first field app so inspectors can complete audits where there is no signal, with sync when they are back online.
- A rules engine that blocks a non-compliant filing at submission.
- A workflow that converts every recorded defect into a tracked work order.
- A command centre for the agency's statutory work.
How it works
- Audits are completed in the field app, offline if needed, and synced later.
- A filing that fails the rules engine cannot be submitted.
- Each defect recorded during an audit becomes a work order that is tracked to closure.
What changed
- Non-compliant filings are blocked at submission.
- Every defect becomes a tracked work order.
Stack
If this looks like your situation
Tell us where it hurts. We will say whether it is worth automating, and what we would build first.
Related work

Open source, MCP
2025 to present
AI assistants had no reliable way to read food regulations across markets
Works with Claude, ChatGPT, Cursor, Windsurf and GitHub Copilot, and is live at mcp-food-regulatory.vercel.app.
17markets covered
Created and maintained by Ansuj Joshi, open source, 2025 to present

Agriva, real estate
2026
Buyers could not search farmland the way they talk about it
Chat is now the primary way buyers search on agriva.ai, and every answer is grounded in listing data rather than guessed.
Built by Vyzma. Led by Ansuj.

Agriva, real estate
2026
Brokers pasted WhatsApp text and phone photos, and that went live
Listings go live in English and local languages from one structured call, cheaper and more consistent than a call per language.
Built by Vyzma. Led by Ansuj.