Publish the bad news too
A missed gate, a failed pilot, a delayed report — members hear it from us first, in plain language, before it becomes a discovery.
Thesara AI / About
A company built the way an arch is built.
Stone by stone, from both sides, closing at the keystone — and standing because every part carries load. Thesara AI is a long-horizon software company, not a platform play waiting to be acquired.
Mission
Put an organisation's knowledge to work.
Thesara AI exists to rebuild the software that industries run on so that it does the work rather than recording it — and to make that capability reachable by organisations that are not large enough to build it themselves.
That means two commitments we will not trade away. The first is depth: seven industries, understood properly, rather than a generic product sold everywhere. The second is proof: nothing ships on a demonstration. Every engagement is measured against a number agreed in advance, and reported as it came out.
We measure ourselves on whether a mid-sized hospital, a mill two countries from its buyer, or a school system with no IT department can run the same class of software as a company with a thousand engineers — at a price that makes sense where they are.
Principles
Six rules we run the company by.
Build the boring layer first
Identity, permissions, audit and residency come before features. Every shortcut taken there is repaid with interest for a decade.
The model never decides alone
Where a decision affects someone's health, employment, education or money, a person approves it and the reasoning is logged.
Agree the number first
No engagement begins without a written success measure. A pilot without one cannot succeed, because nobody can say what success was.
Fixes go into the platform
Never into a client fork. It is the discipline that keeps a product company from quietly becoming a consultancy.
Sequence, don't sprint
Seven suites, in waves, each funded by the last. Building everything at once is how software companies run out of money.
The long view
Where this goes after the seventh suite.
The plan in front of members today is software: the platform, the seven industry suites, and the services practice that delivers them. That is a full agenda for several years and we are not distracted from it.
Beyond it, two directions are visible. The first is breadth — logistics, insurance, legal and public administration, each reachable at a fraction of the cost of the first suite because the platform already exists. The second is depth of ownership: as workloads become large, predictable and contracted, operating our own regional compute becomes economically sensible, improving both margin and the residency guarantees we can offer regulated clients.
Neither is funded by the current plan, and neither will be started quietly. Any move into owned infrastructure would come to members as its own decision, with its own capital plan and its own risk analysis.
Why we say this out loud. A company that hides its ambitions looks smaller than it is; a company that funds them without asking looks worse. Publishing the roadmap and stating plainly what is not being paid for is the version that survives an audit.
Contact