velopersCase studies.
The problem each business had, the AI we built for it, and how it helps them scale.
An AI recruiter that interviews every applicant
Airbase is recruiting software with an AI recruiter called Sage at its centre. We designed and built it. Sage screens, interviews and ranks every applicant, so recruiters spend their time on the people worth meeting.
Read the case study- The problem
- A recruiter can only read so many resumes and hold so many first-round calls in a week.
- What we built
- An AI recruiter that runs the pipeline. First-round interviews with no recruiter on the call. Rules written in plain English.
- What changed
- Recruiters only meet the best.
A paid product where AI marks every answer
Dutch Exams is a practice platform for the Dutch inburgering and Nt2 exams. We built it as a subscription product. Learners take timed tests in every skill, and AI marks their speaking and writing the way an examiner would.
Read the case study- The problem
- Reading and listening can be marked by a quiz.
- What we built
- Timed practice in every skill. AI that marks speaking and writing. Free and premium plans.
- What changed
- A live subscription product, with AI doing the marking.
From a script to a finished video, without a film crew
scnz is an AI video studio we designed and built. You write a script, and an AI director plans and produces the video: the visuals, the voiceover and a cut you can edit.
Read the case study- The problem
- Video is how most businesses now need to explain and sell, and it is the most expensive thing they make.
- What we built
- An AI director. One consistent look. Voiceover without a microphone.
- What changed
- Start from a finished cut, not a blank timeline.
An AI agent that onboards every new customer
Pavago matches businesses with people from its own talent pool. We built the agent that meets each new customer. It talks with them to understand what they need, then recommends the people who fit.
Read the case study- The problem
- Every new customer started with an intake call.
- What we built
- A guided conversation. Matches from Pavago’s own pool. A clear brief for the team.
- What changed
- Every customer understood before anyone picks up the phone.
An AI assistant that works inside your own software
agentside is the engine behind the assistant in Airbase, and we made it open source. It lets an AI assistant do real tasks in any software product through that product’s own API, as the person who is logged in.
Read the case study- The problem
- Most AI assistants added to software can only chat.
- What we built
- It acts through your API. New tasks taught in plain writing. Any AI model.
- What changed
- In production inside Airbase, and free for anyone to use.