Projects
FersmithAI: Personalized AI/ML Learning Platform In Progress
A full-stack SaaS platform delivering personalized AI/ML learning roadmaps for technical professionals. Built solo with Next.js 14, Supabase (PostgreSQL + Auth), and Tailwind CSS. Features a 140-question adaptive assessment engine, dynamic skill graph with real-time progress tracking, and Google OAuth. Currently live as a waitlist — subscription model coming soon.
Visit fersmith.ai
SkyAnalyst-AI: Cabin Operations Intelligence Framework
A modular NLP framework that transforms unstructured flight cabin reports into structured operational insights using RAG and abstractive summarization. Built on ChromaDB vector storage and all-MiniLM-L6-v2 embeddings for semantic retrieval across 2,500+ synthetic incident reports — enabling meaning-based search that goes beyond keyword matching. Summarization is powered by DistilBART-CNN, delivering 4x faster inference than standard BART with negligible quality loss. Includes an exploratory analytics dashboard with sentiment intensity scoring and statistical visualizations across 5 incident categories and 22 flight routes.
View on GitHub
Transformer-Based Summarization & Evaluation
Fine-tuned and evaluated multiple transformer models (BERT, T5, mT5) for Turkish children’s literature. Integrated Flask API, and Stable Diffusion to generate visual illustrations. 90% of users found the visuals well-aligned with the story content.
View on GitHub
Meat Quality Detection
Developed a TensorFlow-based image classification system to detect spoiled meat. Applied data augmentation and evaluated model performance with accuracy and loss metrics.
View on GitHub