Hi, I'm Taher Nouira
Software Engineering student · AI, GenAI & Computer Vision
I build intelligent systems that bridge the gap between cutting-edge research and real-world applications. Currently exploring retrieval-augmented generation, multimodal AI, and the intersection of language and vision.
What I Work On
LLMs & RAG
Building retrieval-augmented systems that ground language models in real data, exploring prompt engineering and fine-tuning techniques.
Computer Vision
Experimenting with multimodal models, image understanding, and the integration of visual and textual information.
ML Systems
Designing end-to-end pipelines from data processing to deployment, focusing on scalability and real-world usability.
Featured Projects
View all projects →
MiniGPT
FeaturedA fully custom GPT implementation from first principles — including tokenizer, Transformer, training, API, MLOps tracking, and an interactive web UI.
RAG Hotel Recommender
FeaturedA dual-module system combining fake review detection using BERT and personalized hotel recommendations via Retrieval-Augmented Generation (RAG) with Gemini.
Serendipity Engine
FeaturedAn AI system that generates relevant yet surprising ideas by linking concepts across domains using LLM embeddings and semantic search.
PENELOPE
FeaturedA rule-based prompt sanitization engine designed to detect, neutralize, and rewrite unsafe or adversarial user inputs before reaching LLM backends.
Limited details (NDA)
TINA
FeaturedA visual document classification system that identifies Energy Performance Certificates (EPCs) and their country of origin.
Limited details (NDA)