Design, extend and connect Dotwork — in a conversation.
“How fast can we make this fit the way we actually work?”
Dotwork is AI-native all the way down, so the platform builds with you. Extend the ontology by describing your business. Connect a system by naming it. Build a skill that acts and weighs tradeoffs by talking it through. Every studio has two ways in — tell Dot what you need, or build it by hand — with the same result, the same flexibility and the same guardrails.
You → Dot
“Connect our Jira and bring in epics with their status and owner.”
What changes
What changes when the platform builds with you.
Months of workshops to map the operating model
A working model sketched in a conversation, refined by hand
An integration project with tickets and specs
A system connected by describing it — tested before it saves
An admin screen for every new field
The field added the moment the conversation needs it
Automation scripts only their author can read
Skills written in plain language that anyone allowed can run
Shape the model of how your enterprise runs
Ontology Studio
Model your business in a conversation.
Tell Dot how your company works and watch the model take shape — the objects you run on, how they relate, the lists and statuses they use. Start from a quick sketch or a full working session, then refine by hand in the Model Designer before anything reaches your live site.
- Build a model by conversation, from a quick start to a full session
- Add objects, fields and relationships to a live site by asking
- Versions, side-by-side comparison and validation before you deploy
In a conversation
We fund value streams, not projects. Each value stream has products, and products have teams and budgets.
- ✓Proposed Value Stream, Product, Team and Budget objects
- ✓Linked them: value streams own products, products have teams
- ✓Checked the model and fixed two gaps
- ✓Ready for you to review in the Model Designer
Illustrative conversation
Or build it by hand
- Objects, fields and relationships
- Select lists and status labels
- Spaces and lifecycles
- Published versions and comparison
- Reverse-engineer a model from a live site
- Validate, then deploy
Guardrails
Site admins only. Changes can pause for approval, and every model is validated before it deploys.
Connect and map the systems that feed the model
Integration Studio
Connect a system by describing it.
Ask Dot to connect a system and it finds the provider, gives you a secure sign-in card, reads the API documentation and drafts the requests — testing each one before it saves. If your model has nowhere to put the data, Dot proposes the objects and fields in the same conversation.
- A large provider catalog, plus any API you point it at
- Requests drafted and tested before anything is saved
- Credentials entered in a secure card — never in the chat
In a conversation
Connect our Jira and bring in epics with their status and owner.
- ✓Found Jira in the catalog and opened a secure sign-in
- ✓Read the API documentation and drafted the request
- ✓Tested it on live data before saving
- ✓No object fit — proposed an Epic type with status and owner
Illustrative conversation
Or build it by hand
- Connections and sign-in methods
- Requests with parameters and preview
- Values and lists built from responses
- Smart links to records in the source
- Schedules and run history
- Health checks and where each value is used
Guardrails
Anyone can preview and test; admins connect and schedule. Expressions that fail or return nothing are refused, not saved.
Build apps and views on the live model
Apps Studio
Views, readouts and apps on one set of numbers.
Build the rooms and views each team needs from a library of view types, all reading the same live model. Ask Dot for a readout and it drafts the report with its charts. In scenario planning, Dot builds interactive apps, pivots and decks you can take into the meeting.
- Rooms and views for every team, on live data
- Readouts drafted by Dot, with charts
- Interactive apps and decks in scenario planning
In a conversation
Build me a readout of Q4 spend against plan by value stream for Thursday’s review.
- ✓Pulled plan and actuals for each value stream
- ✓Drafted the readout with a variance chart
- ✓Flagged the two value streams furthest off plan
- ✓Saved it for you to edit and share
Illustrative conversation
Or build it by hand
- Rooms for each team and review
- A library of view types
- Boards, tables, charts and timelines
- Readouts with charts
- Scenario apps, pivots and decks
Guardrails
Interactive apps never write on their own — any change is previewed and approved by the person using it.
Surveys and the context AI should know
Context Studio
Capture what your systems don’t hold.
Some of what runs a company lives in people’s heads. Run surveys — assessments, attestations, intake and pulse checks — whose approved answers write back to the model. And tell Dot what it should always know about a site, a space, a kind of record or a skill.
- Surveys whose approved answers write back to the model
- Context notes attached where they apply
- Context that never widens anyone’s access
In a conversation
Remember that in this space, “committed” means funded and staffed — not just approved.
- ✓Saved a note to this space’s context
- ✓Applied it to every request made in the space
- ✓Kept it visible and editable by the space’s owners
Illustrative conversation
Or build it by hand
- Survey templates with questions, scales and scoring
- Campaigns, reviewers and calibration
- Approved results published to fields
- Context notes for sites, spaces, records and skills
Guardrails
Survey templates are versioned and published, with separation of duties between respondents and reviewers.
Skills and decisions that do the work
Actions Studio
Build skills that act and weigh tradeoffs — by talking them through.
Describe a job and Dot investigates your data, drafts the steps and saves a skill your teams can run on request or on a schedule. Define decisions — judgments Dot makes by weighing evidence across the model — and keep people on every change that matters.
- Skills built and refined with Dot
- Decisions that weigh evidence across the model
- Changes proposed and routed to the right approver
In a conversation
Each week, flag initiatives whose spend is ahead of their progress and propose which to slow down.
- ✓Checked which fields hold spend and progress
- ✓Drafted the steps: compare, rank, explain
- ✓Set it to propose changes, not make them
- ✓Routed approvals to each portfolio owner
Illustrative conversation
Or build it by hand
- Skill steps, questions and signals
- Context the skill always reads
- Run on request or on a schedule
- Decisions with evidence rules and budgets
- Approvals routed to a named person
Guardrails
Skills run with the permissions of the person who runs or schedules them, and can propose instead of change.
One conversation, every studio
Studios work together, because they share one model.
A single request can cross studios. Connecting a system can extend the model, fill it with data and feed a readout — without leaving the conversation.
- 1 · Integration Studio
“Connect our Jira and bring in epics.”
- 2 · Ontology Studio
No object fits — Dot proposes an Epic type with status and owner.
- 3 · Integration Studio
Epics fill the new type, on a schedule you set.
- 4 · Apps Studio
A readout of epics by value stream, ready for the review.
- 5 · Actions Studio
A weekly skill that flags epics with no owner.
Fast, and still governed
Building in a conversation doesn’t mean building without rules.
Two ways in, one result
Whatever Dot builds lands in the same editors you’d use by hand, so you can always see and change it.
Admins own the structure
Changes to the model and to connections need an admin, whether they come from a conversation or a form.
Approval when it matters
Changes can pause for human review, and skills can propose instead of change.
Tested before saved
Requests are tried on live data and models are validated before they deploy.
Credentials stay out of chat
Sign-in happens in a secure card, never in the conversation.
A record of what changed
Published model versions you can compare, and a change log for the site.
The model underneath
Every module makes the next one smarter.
Steer, Fund & Optimize and Govern & Deliver all read and write the same Enterprise Operating Model. The first module you license builds the model. Every module after it adds context instead of another silo — and Dotwork AI reads all of it, stopping where the decision is yours.
See the Enterprise Operating ModelBe Decisive.Build Fast.Connect Everything.
Start ConnectingSolutions
Platform
Resources
© 2025-2026 Dotwork, Inc. All rights reserved.