Done is a VC-backed startup out of Stockholm with the ambition to become the easiest way to hire craftsmen for your home projects. Our tech-stack includes Flutter, Node.js, Firebase and WebRTC. We are now looking for talented thesis workers for the autumn.

What we offer is the chance to work with cutting edge tech, high degrees of freedom, a strong engineering culture and the opportunity to post your work as open source (some of the proposed projects).

We accept single applications as well as pairs for some of the bigger projects (like within AR). Please provide a transcript of your current university grades as part of your application, as well as links to anything you've engineered previously and what project you are interested in pursuing.

Fully remote is accepted for top students but preferably you are able to come to the Stockholm office occasionally. Equipment will be provided during the thesis period if needed.

You will be tutored by our VP of Engineering Frederik as well as our CTO and co-founder Alek. The thesis will be compensated with a standard rate of 30 000 SEK upon completion.

Looking forward to your application

Project proposals

Using the Apple RoomPlan API for Estimating Craftsmen Work in an AR Experience for Home Owners.

RoomPlan is a new Swift API that utilizes the camera and LiDAR Scanner on iPhone and iPad to create a 3D floor plan of a room, including key characteristics such as dimensions and types of furniture. This project involves creating an AR experience for estimating flooring & painting work, optionally highlighting other items in the home which can be replaced by a craftsman. We expect to release this at launch of iOS 16 together with surrounding PR. 

Review of Technologies & Architectures for Live-Transcription of WebRTC Calls

The Done platform & Flutter app currently hosts a custom WebRTC-based video calling solution. We want to provide customers with transcripts of their calls so customers can recall what was said in a previous call. The project is to evaluate different solutions for transcribing the calls, live on device or server-side. 

Sentiment Analysis of Customer Chats for Probability Prediction of Sales Closure

During the quotation process, customers chat with their matched craftsmen before committing to their project and accepting or declining their quote received from the craftsman. This project aims to train an ML model to predict a positive or negative outcome based on sentiment analysis of chat contents. The results would be used by our support team to prioritize customers.

Multi-platform Firestore Schema Generation from Typescript Model Definitions

At Done we use Firestore extensively, but it's hard to keep data models in sync between our Flutter app and our Firebase Cloud Functions. This thesis aims to create an open-source project which generate Dart (& other) Firestore models based off Typescript model definitions, similar to how GraphQL client generation works. So that all our cloud functions  

Integration Testing of Multi-Role Flutter Apps Across Mobile and Desktop

Evaluate & implement frameworks and methods for integration testing the Done Flutter app across iOS, Android and Web with multiple form-factors and user roles. 

Other projects

Do you have another idea for a thesis project that would benefit Done? Feel free to submit a proposal.