Skip to content

What it is built on

Four opinions, held firmly

A tool is its decisions. These are ours, and each one is visible in the product rather than only on this page.

A story is structured data

Characters, items, and traits are typed records with fields you define — not names in a document you hope stay spelled the same way. Rename one and every mention of it follows. Most narrative tooling treats structure as an export format; here it is the thing you author.

The tool should not own your schema

You author your own node types and definitions rather than bending your world to fit someone else's idea of what a character is. A project about a legal drama and a project about a dungeon crawl should not be forced through the same fields.

What the writer authors is what the game runs

The Unity and Unreal plugins generate types from your definitions and sync continuously, so the records a writer edits are the records the build reads. No exporting, no re-entering, no copy-pasting identifiers between two tabs.

Automation belongs on the clerical work

The AI converts what you have already written into project data and answers questions about your own story. It can generate freely if you ask it to, but that is not where the effort went — the effort went into deleting the retyping.

Who builds it

An engineer's tool for other people's stories

C

Carlo Martin

Founder & architect

Designs and builds TaleShard end to end — the editor, the collaborative backend, and the infrastructure underneath it.

TaleShard is not the work of a narrative department that wanted better software. It is built by a developer, and that shows in what it is opinionated about — types, schemas, references that stay correct, and getting data into an engine intact.

It has deliberately no opinion about how you should write. There is no house structure to conform to, no recommended act count, no assistant nudging your prose toward the average of its training set. The craft is yours. The bookkeeping around it is the part we think software should have solved by now.

That division is the whole design, and it is why the parts of this product that get the most attention are the ones nobody writes blog posts about: typed fields, a real undo history, an import that does not mangle your dialogue.

Where it stands

TaleShard is in development and access is by waitlist, opening in waves. The roadmap lists what is already built and what comes next — it is more useful than a launch date we would only have to move.