Twin · pre-awardGRANTAPPLICATION
Twin is the workspace that makes AI fit for grant work. Context that travels with you, statements that point back to their source, client data that stays in the EU, an audit trail that is standard.
For grant teams that want to stay in control of the whole work process, from spotting an opportunity to submitting a grant application.
How an application moves through Twin.
From spotting to submitted application, with you in control.
Spot
Schemes and calls come in with their deadlines and conditions, so a project idea becomes an opportunity before the call closes.
Qualify
One place for all project documents: intake, quote, earlier applications and so on. That context shows early whether a project is worth pursuing, and what you record here becomes the basis of your settlement file after the award decision.
Match
Twin sets the project against the scheme's requirements and automatically outlines the application in that structure.

Draft
AI writes with the applicant's project context, within the character limit the scheme sets. You review and improve as a team in real time, with an audit trail.

Submit
Automatic quality control before submission. The same criteria return later in reporting and final settlement.
Section 3 · Innovative approach
What Twin adds to a language model.
For anyone already using ChatGPT or Claude and wondering where Twin makes the difference.
AI with project context
Twin works with your documents, earlier applications, writing guidelines and scheme requirements.
Hallucination control
Every AI statement points back to a source line. If that source is missing, the sentence gets flagged.
Automatic quality control
Completeness, consistency, compliance with scheme requirements and tone, checked automatically before submission.
Collaboration with an audit trail
Work simultaneously with colleagues, experts and clients. Every change and AI call is logged.
Learning within your organisation
Twin reuses successful wording and structures, exclusively within your organisation.
Every scheme, the right approach
From WBSO, the Dutch R&D tax credit, to Horizon Europe. Structure, tone and evidence adapt automatically.
Infrastructure that matches your responsibility as a professional.
Using AI in grant work calls for infrastructure that accounts for what happens to client data.
Read the full data policyEU-hosted, ISO 27001
Hetzner cluster in Helsinki. No US cloud, no transfer outside the EU unless you choose it yourself.
Client data stays out of the models
We do not change model weights based on Twin input. Strict multi-tenant isolation in every layer.
Strongest model, as standard
Frontier models for the best output. Optionally fully EU-hosted or local on Twin infrastructure in Finland.
Before AI starts, Twin knows whether the input is good enough.
Per section, based on the scheme's requirements. Per criterion a score, a reason, and a marker for what is missing.
- Completeness tested against scheme requirements
- Substantiation per criterion, in plain language
- Points out exactly what needs adding
Background describes the problem or the opportunity
70 / 100The problem is there, but not why it has to happen now. Add what goes wrong if the project starts a year later.
Objective links to the scheme
160 / 200The objective is quantitative and refers to the scheme. An explicit link to the assessment framework would close this criterion.
Core activities are described concisely
45 / 100The activities are mentioned implicitly, not structured as work packages. Division of tasks and coherence are missing.
Result is concrete and measurable
55 / 100There is a direction, not a measurable end result. Name the tangible outcome and how you will measure it.
Summary is suitable for publication
35 / 100The summary contains details you do not want made public. Twin marks which passages need anonymising.
Twin in action.
In just four minutes, from uploading sources to a validated submission. Watch the video! The video is in Dutch.
Every tool has its place.
Twin is built as a complete platform for the work process from spotting to submitting. An honest comparison.
Twin
Everything in one place, for grant work.
- AI with context
- Source links
- Real-time collaboration
- Audit trail · ISO 27001
- EU-hosted
- Scheme-specific output
CRM + project tool
Keeps track of the schedule.
- Deadlines and tasks
- View of the pipeline
- Scheme knowledge
- Dossier building
- Quality check
ChatGPT / Copilot
First draft and brainstorming.
- Fast text generation
- Broadly usable
- Scheme knowledge
- Collaboration
- Audit trail
Grant databases
Where the search starts.
- Matching schemes
- Deadline overview
- Process support
- Templates
- Collaboration
Word + SharePoint
Familiar, but scattered.
- Full control
- Version control
- AI with context
- Templates
- Quality check
Get in touch with us.
Call us for more information, advice or a trial account. We are happy to help.

“Writing is faster, but the real gain comes before that: Twin forces the evidence to be complete before we submit. For every question from the assessor, we have the answer right there.”
Martijn
Cata Subsidieadvies
