Traineeship title: Developer
Reference: N154469/20/MADEIRARegion
MadeiraCountry
Portugal
Traineeship start date:
01/07/2026Traineeship end date:
30/11/2026Traineeship Duration:
Approximately 5 monthsDeadline
30/04/2026Work Schedule
Work Schedule is from monday to friday, 9:00 to 18:00 with 1 hour of lunch, however this schedule is flexible.Location
FunchalTraineeship Area:
IT and TelecommunicationsPeriod and duration of the language course:
The Portuguese language course takes place during the first month of the internship (1 June to 30 June 2026).Includes Trips
NoDescription of the tasks to be undertaken by the trainee(s):
Develop an immersive AR application using Unity where the core engine relies on on-device computer vision (like TensorFlow Lite or Niantic ARDK) to perform Object Detection and Classification in real-time. Instead of looking for pre-defined 2D image targets, the camera feed is processed by a neural network trained to identify 3D objects (e.g., a car tire, a leaf). Upon successful classification, the app acts as a database lookup, querying a backend chemical library to map the recognized object to its primary chemical components (e.g., "Car Tire": Vulcanized Rubber/Polyisoprene). Finally, the system instantiates a 3D molecular visualization anchored to
that object in world-space, providing interactive data on bond types, atomic structure, or material properties.
Profile of the Trainee
Qualifications
Bachelor's or Master's degree in Computer Engineering
Proficiency in training or fine-tuning Convolutional Neural Networks (CNNs) to recognize specific 3D objects in real-time camera feeds. Experience compressing models to run on mobile processors (e.g., converting PyTorch/TensorFlow models to TensorFlow Lite or Unity Sentis) without causing latency or overheating. Ability to implement Instance Segmentation, allowing the app to not only recognize an object but also delineate its exact edges to anchor the AR model precisely on it.
ARCore (Android) or ARKit (iOS) to understand the physical geometry of the room (planes, points, lighting), with solid grasp of vectors, matrices, and quaternions for manipulating 3D chemical models in relation to the camera's perspective.
Experience mapping real-world lighting onto virtual objects so the 3D molecule looks like it is actually in the room. Ability to design or integrate a RESTful API that links object IDs (e.g., "Steel_Beam") to a chemical database. High proficiency in shaders and materials to visualize complex atomic structures (e.g., Lewis structures, space-filling models).
Coding intuitive interactions, such as allowing a student to "tap" a virtual atom to see its electrons or "drag" a reagent towards an object to initiate a simulation.
Personal and Social Skills
Autonomous, team working, good listener, objective oriented
Language Skills
English
Information Technology (IT) Skills
Unity/C#, ARCore/ARKit, 3D Rendering. ML EngineerObject Detection, TensorFlow Lite,
Computer Vision. SQL/NoSQL databases, Python.
Other Skills
Other Skills
Interns must bring their own laptop and have the Unity programme installed.