Gunly is an AI bid assistant. It reads every page of the tender, maps every requirement, drafts your answers from your own past bids, and scores them the way the evaluation committee will — so you do half the work and submit stronger bids.
Large firms have tender teams. Gunly gives every company an AI one — each step of the bid handled by a specialised agent, so a 60-person contractor can bid with the same precision as a multinational.
An agent scans TenderNed and TED every day and matches new notices against your certifications, references, turnover and region.
Gunly reads the tender's own evaluation method and plays the committee. Every criterion gets a predicted score and specific, actionable feedback, so you know exactly where the points are hiding.
Fix it, rescore it, repeat. Your team decides what goes in — Gunly never submits anything on its own.
Gunly runs on Claude, Anthropic’s AI model — chosen because it can read entire tender files at once, reason carefully across annexes, and write clear, precise prose. On top of it, we built agents that know how public procurement works.
Long-context AI takes in hundreds of pages in one go, so requirements in annex 7 are linked to the criteria on page 3.
Every extracted requirement and every claim in a draft links back to the page or reference it came from. Check it in one click.
Your past bids, references and award reports form a private knowledge base. The more you bid, the sharper your drafts and scores get.
The AI never submits anything. Your team reviews every answer and stays fully in control of the bid.
Public contracts should go to the best offer, not to whoever can afford a full-time tender team. Yet every week, skilled contractors, IT firms and care providers skip tenders they could win — simply because the paperwork is too much.
We are a small team from the Netherlands, building Gunly together with our first customers. Every feature you see here started as a conversation with someone who writes bids for a living.
Beta teams help decide what we build next. Your feedback goes straight to the people writing the code.
All data is stored and processed in EU data centers, designed with GDPR in mind.
Each customer's bids and references live in a separate, encrypted workspace. Never shared, never mixed.
Gunly drafts and advises. Your team reviews, decides and submits — every time.
Beta teams use Gunly for free until public launch. After that, you keep your founding price — as long as you stay with us.
You get full access for free, a direct line to the founders, and a say in the roadmap. In return, we ask for honest feedback — a short call every few weeks.
Gunly is an assistant, not a ghostwriter. Your team owns every answer and makes every decision. Always check the tender's own rules on the use of AI.
We start with TenderNed in the Netherlands and TED for the whole EU, with more national platforms on the roadmap.
No. Your documents are only used inside your own workspace to improve your drafts and scores. They are never used to train AI models and never shared with other customers.
We onboard a few new teams every week, personally. Leave your work email and we’ll reach out to plan your first tender together.