Reviewer intelligence for grant writers

Know where a reviewer will push back — before you submit.

Whetmark reads your proposal like a real evaluator, scores each criterion against a private corpus of genuine reviewer criticism, and marks up the fixes in your Word document. All on your own machine.

Local · nothing leaves your machine On-device AI EU-hosted
● Mock evaluator · pre-submission read

Reviewer Risk Score

Where an evaluator is most likely to object — lower is safer.
Excellence
Moderate
Impact
Lower risk
Implementation
High risk
Trained on 30 real grant proposals and the official evaluator reports they received — analysed on your machine, never uploaded.
The gap Whetmark closes

You write for the call. Reviewers score against everything they've rejected before.

Most tools polish your grammar. Whetmark does the one thing that changes outcomes: it anticipates the evaluator's objections while you can still fix them — because we studied real grant proposals alongside the official reports their evaluators wrote, and taught the tool to spot the same problems in yours.

Predict the pushback

Every section is scored 0–10 against Excellence, Impact and Implementation — the three criteria a Horizon or national panel actually uses — and shown the exact past criticism your text most resembles.

Fix it in your document

Accept the suggestions you want and Whetmark writes them straight into your proposal as Word tracked changes — not a list to retype. You stay in control of every edit.

Iterate until the risk drops

Re-score after each pass and watch the gauges fall version over version. You submit only once the evaluator's likely objections are answered.

The non-negotiable

Your unpublished proposal never leaves your machine.

Whetmark runs its AI locally, on your own instance. No OpenAI, no Gemini, no cloud model ever sees your ideas, your consortium, or your budget. The confidentiality that a competitive proposal demands is built into the architecture — not a setting you have to trust.

  • On-device inference — the models run beside your document, not in someone else's datacentre.
  • Self-hosted on a private EU instance, dedicated to you.
  • The reviewer corpus is matched locally; identities of past proposals stay hidden.
Why local matters for grants

A grant proposal is your most sensitive unpublished work — novel science, a costed plan, a named consortium. The moment it touches a third-party AI it can be logged, cached, or trained on. Whetmark removes that risk entirely: there is no third party.

Built for Horizon Europe, ERC, MSCA, FWO & national calls.
A method, not just a tool

Six phases, from “should we bid?” to submission.

Whetmark is organised around the workflow in the companion handbook — so the tool and the method reinforce each other.

A

Qualify & set up

Decide to bid; plan the deadline.

B

Intake & safety

Bring documents in, locally.

C

Decode & frame

Turn the call into a map.

D

Design & draft

Impact pathway and the work.

E

Review & iterate

The mock-evaluator pass.

F

Finalize & submit

Tracked changes, final checks.

The companion handbook

Grant Writing with AI

Professional Consortium Edition · 18 recipes

The playbook Whetmark operationalises — the practical method behind the phases, written for professional grant writers and research offices.

Method + machine

The book teaches the craft. Whetmark runs it.

Every recipe in the handbook has a home in the workspace — so a team learns one method and executes it in one place. New clients get the book as part of onboarding.

Private instances

One dedicated, EU-hosted instance per client.

We provision and manage a private instance for you — no installation, no cloud AI. Pricing is per instance, invoiced by Scientific Network Management S.L. (indicative tiers — final quote on request.)

Starter

on request
Solo consultants & single researchers.
  • Private managed instance
  • Reviewer risk scoring
  • Tracked-changes export
  • The companion handbook
Request a quote
Most popular

Professional

on request
Grant writers & research-office teams.
  • Everything in Starter
  • Full multi-agent review sweep
  • Reviewer-intelligence corpus
  • Version-over-version risk tracking
  • Priority support
Request a demo

Consortium

on request
Universities & multi-partner bids.
  • Everything in Professional
  • Multiple seats & projects
  • Dedicated onboarding
  • Custom corpus & calls
Talk to us
Try it on a real call

See your next proposal through a reviewer's eyes.

Book a walkthrough on the live demo, or ask for a private instance for your team. We'll reply within one working day.

Request a demo Open the live demo