Project overview:

• Development of an AI-powered visual learning application that converts complex or abstract words into meaningful images to help children with dyslexia or language learning challenges.
• Position open for 2 trainees focused on building a multimodal system combining NLP and image-generation/retrieval to adapt visuals to each learner.

Objectives & scope:

• Automatic detection of complex vocabulary using NLP techniques to identify words that require visual support.
• AI-based image association using CLIP / BLIP or image retrieval APIs to generate or fetch appropriate visual representations.
• Design of a visual learning interface including word cards, image previews, and pronunciation playback to support memorization.
• Creation of a custom dictionary module targeted at teachers, parents and speech therapists, plus an optional offline mode for classroom use.

Responsibilities / Tasks:

• Implement NLP pipelines for difficulty detection and vocabulary analysis (tokenization, lexical complexity metrics, context analysis).
• Integrate multimodal models (CLIP/BLIP or HuggingFace models) and/or image retrieval APIs to map words to images and rank candidate visuals.
• Develop backend services (FastAPI / Nest.js) and data storage (MongoDB) for user data, dictionaries and offline content synchronization.
• Implement front-end features in React Native / Next.js: word cards, image previews, audio pronunciation and teacher/therapist tools.

Required profile & expectations:

• Profile: Software Engineer trainee with interest/experience in NLP and multimodal AI; ability to work on full-stack components.
• Duration: 4 months of full-time internship with hands-on development and possible iterations with teachers/therapists.
• Soft skills: collaboration with educators, iterative testing with learners, attention to accessibility and usability for children.

Technologies & tools:

• Python with CLIP / BLIP models and HuggingFace ecosystem for multimodal tasks.
• Backend frameworks: FastAPI or Nest.js; Frontend: React Native and/or Next.js for mobile/web interfaces.
• Database and deployment: MongoDB, Docker, Git/GitHub for versioning and CI workflows.

How to apply:

• Apply via the company website: https://www.readdlytech.com
• Or send your application by email to contact@readdlytech.com including CV, motivation letter and link to any relevant projects or code repositories.