LLM Personalized Music Assistant for Cognitive Tasks
Built an LLM-powered web prototype and designed a study of personalized background music, attention, memory, and planning.
Read the storySeyedehsahar Fatemi Abhari
AI, human–AI interaction and software development
I’m Sahar Fatemi, a master’s student in Computer Science and Engineering at the University of Oulu. My interests centre on AI, human–AI interaction and software development. I build research prototypes, machine-learning pipelines and web applications, including an LLM-powered music assistant prepared for a study of personalized background music and cognitive performance.
Selected projects
These projects bring together study design, data preparation, model comparison and software implementation—from personalized music and visual emotion recognition to a learning platform, earthquake classification, recipe recommendations and retinal imaging.
Learnia – Learning Platform
Developed a microservices learning platform that processes study files into summaries, key concepts, and flashcards.
Visual Emotion Recognition on MELD: EmotionCLIP vs. CLIP
Built a video-preprocessing and annotation pipeline and collaborated on comparing EmotionCLIP and CLIP for seven-class emotion recognition.
USGS Earthquake Magnitude Classification Pipeline
Built a Spark pipeline for USGS earthquake data and compared models for multiclass magnitude classification.
Read the storyRecipe Recommendation LLM Fine Tuning with LoRA + RAG
Fine-tuned TinyLlama for recipe recommendations, integrated retrieval, and compared fine-tuned and RAG outputs.
Read the storyMulti Label Retinal Disease Detection via Transfer Learning
Built a multi-label retinal imaging pipeline and fine-tuned CNNs for DR, Glaucoma, and AMD.
Read the storyBackground and contact
See my background for education, research and development experience, and teaching.
Get in touch to discuss my work, research opportunities, internships or employment.