⁨Companies⁩⁨Tawtheiq⁩⁨Enterprise governance · 2024 — 2026⁩

⁨Every decision in the room, on the record.

⁨Tawtheiq treats a meeting as a lifecycle rather than an event — agenda, attendance, discussion, vote, action, approval — and keeps every step attributable. Two years of Flutter work, with an AI copilot that operates the system by chat or voice.⁩

⁨Meetings run⁩
⁨Approval time⁩
⁨Organisations live⁩

Figures pending. This page argues from what shipped, because that is what the record holds. Before launch, ask the client for two numbers — meetings run through the platform, and approval time before and after. Nothing is published here until one of them is confirmed.⁩

⁨App screenshot to come⁩⁨The meeting record — ideally the agenda view and a decision under vote, side by side.⁩⁨16:9 · source: partner portfolio captures, pending client clearance⁩
⁨Placeholder — real captures exist in the partner's portfolio and need clearing before they go up.⁩
⁨— The question⁩

⁨What happens to a decision after the meeting ends?

⁨Minutes get typed up somewhere. Actions get remembered by whoever remembers them. A month later nobody can say who agreed to what, on what date, or whether the thing was ever formally approved. The meeting itself was governed — chaired, minuted, attended by the right people. The record of it was not.⁩

⁨That gap is where most meeting software stops, because most meeting software is a document. It captures the discussion and then hands the outcome back to email. Tawtheiq's premise is that the decision is the artefact, not the notes: a vote taken inside a fixed window, by named roles, producing an action with an owner and an approval chain.⁩

⁨The distinction matters most in the organisations that need it most — committees, boards, public bodies — where "who approved this" is a question with consequences, and the honest answer today is usually a search through someone's inbox.⁩

⁨— What was built⁩

⁨Twelve capabilities, one lifecycle.

⁨Each of these exists because a step in the lifecycle needed it. None of them is a module you switch on separately — an action item cannot exist without the decision it came from, and a decision cannot close without the vote that carried it.⁩

  • Agenda builder. Topics with time allocated to each, so an hour-long meeting is planned as an hour.⁩
  • Attendee roles. Chairperson, reporter, secretary and member — each with its own permissions over the record.⁩
  • Timed decision voting. Accept, reject or abstain inside a fixed window, with an anonymous-ballot option.⁩
  • Action items. Derived from decisions, tracked by completion percentage, with attachments.⁩
  • Multi-level approval. Minutes and actions route through a review chain rather than one signature.⁩
  • Digital signature. Drawn or uploaded, captured against the approval it belongs to.⁩
  • Azure AD sign-in. Plus fingerprint and face recognition on the device.⁩
  • Committees and teams. Each with its own scope, description and permission set.⁩
  • Search and filtering. Across meetings, decisions and actions, with several match modes.⁩
  • Microsoft Teams integration. Meetings created in Teams carry through into the record.⁩
  • Notification preferences. Push, email and SMS set per channel rather than one global switch.⁩
  • Multi-account sync. One person, several organisations, with granular privacy settings.⁩
⁨— The copilot⁩

⁨An assistant that acts, not one that summarises.

⁨Tawtheiq Copilot answers questions about meeting status, creates action items and approves minutes — through chat or a spoken command. The distinction worth naming is that it operates the system rather than describing it. Same permissions, same approval chain, same audit trail as a person tapping the buttons.⁩

⁨That constraint is the interesting engineering. An assistant that only reads is easy to add and easy to ignore. One that writes into a governance record has to be bound by every rule the record already enforces — which is why the copilot sits behind the same role model as everything else, and why an anonymous ballot stays anonymous to it.⁩

⁨Voice and chat are an interface to the workflow, not a summary of it.⁩

Why this is the line that matters. Most "AI in the product" is a chat box bolted to a search index. The test is whether the assistant can complete a task the product exists to do — and be held to the same rules when it does.⁩

⁨— Trust⁩

⁨Who signed it, and how you know.

⁨Sign-in runs through Azure AD, so the app inherits the organisation's own directory rather than keeping a second list of who works there. On the device, fingerprint or face recognition stands between a borrowed phone and a signature.⁩

⁨Approvals are captured with a digital signature, drawn or uploaded, attached to the specific approval it belongs to. Decisions can be put to an anonymous ballot inside a fixed voting window — which means a vote cannot be quietly reopened once it has closed, and that is the property the whole record depends on.⁩

  • ⁨Flutter⁩
  • ⁨Dart⁩
  • ⁨Bloc⁩
  • ⁨Azure AD⁩
  • ⁨REST API⁩
  • ⁨Firebase⁩
  • ⁨AI integration⁩
  • ⁨Biometrics⁩
⁨Role⁩
⁨Mobile app development⁩
⁨Period⁩
2024 — 2026
⁨Platform⁩
⁨Flutter, one codebase, iOS and Android⁩
⁨Availability⁩
⁨Private enterprise deployment — distributed inside the organisation against its own directory, so there is no public store link.⁩
⁨— Contact⁩

⁨Let's keep in touch.

⁨Most of our work starts with a conversation. Tell us what you're working on — or just introduce yourself and we'll do the same. We answer in a day, usually less.⁩

⁨saleem.najjar@umbrella500.com⁩

⁨No form, no funnel — that address reaches us directly.⁩

⁨See what we build for mobile
⁨— Enterprise workflow on mobile?⁩

⁨Approval chains, directory sign-in and an audit trail that holds up — on a phone. That is companies-side work, and the companies page has the rest.⁩

⁨For companies