Vyom builds a single living model of how one specific company works, then runs a workforce of agents from it. What changes per industry is configuration, never code. Everything here is built and verified on reality-grade synthetic companies with known ground truth, which is exactly what makes these worked examples checkable.
There is one engine: it ingests events, builds a graph of variables and weighted causal edges, propagates a change through it, simulates an intervention before anyone commits, and runs agents that watch, reason and act. It does not know what industry it is looking at. What a company brings is configuration, not new software: its variables, its connectors, the signals that move it, the metrics each role watches. If a line of runtime code ever read if industry == “construction”, that would be a bug we would remove. The vocabulary is per company. The math is shared.
How a real company is onboarded: you sign on, and Vyom ingests your data, historical and live, internal and external. On top of an industry backbone it already holds, warm-started by patterns learned across the fleet, it connects your streams and learns your company’s specific structure by watching how it actually behaves, then confirms that structure against real decisions and their outcomes. What you get is your own live causal model of your business, wired to every stream in real time.
With your permission, and never your raw data, what your model learns quietly sharpens the shared model for every company onboarded after you.
The 30-second, describe-it-in-plain-English spin-up is a different thing: it lets anyone explore the engine on any industry with no data at all, and it is how the synthetic worked examples on this page were built. That is a demo of the industry-agnostic engine, not how a real company is onboarded.
Everything below is an instance of the same six modes running on a different company. They do not change by industry. Meridian Construction’s rain-to-cash chain and a hospital’s no-show-to-revenue chain are the same six modes pointed at different variables.
Eight industries built as templates you can load now, each configuration for the same engine. Beyond the eight, any company can be explored from a plain-English description, and the reasoning runs identically across all of them.
| Industry | Connector status | What that means |
|---|---|---|
| Construction | wired | Integrations built for read and, where it makes sense, write. |
| B2B SaaS | wired | Integrations built for read and, where it makes sense, write. |
| Manufacturing | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| 3PL Logistics | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| Healthcare | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| Hospitality | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| Real estate | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| Restaurant groups | modeled | Connector surface specified and reasoned over; live external hookup is the remaining step. |
| Any industry, in ~30s | demo | Explored from a plain-English description, no data required. |
Open each for the full how, why and what. Because the true structure was authored, every claim is checkable against known ground truth.
Everything above is built and verified against reality-grade synthetic companies whose true cause-and-effect structure we authored, so every claim traces back to a number we can check. Vyom is pre-seed and pre-revenue, with zero live customer deployments. Two genuine outstanding items, everywhere. Full status →
The interactive Explore page is the hands-on version: switch companies and fire a signal through the model yourself. Or start the conversation.
Request access