Medical RAG QA app using Meditron 7B, Qdrant, and PubMedBERT embeddings — evidence-grounded clinical Q&A over medical literature via a Chainlit interface.
"}},{"node":{"frontmatter":{"title":"Ads-LLM","tech":["LLM","Llava","Supabase","Keybert","Vertex AI"],"github":"https://github.com/otmane-elbourki/ads-llm","external":"https://ads-llm.com","order":2,"date":"2024-01-01"},"html":"Implemented a program for matching articles with ads, utilizing KeyBERT to pinpoint relevant article keywords and a custom LLAVA deployed on Vertex AI to extract ad-descriptive keywords. Ensured sensitivity and disallowed keyword compliance during the matching process while storing keywords efficiently on Supabase.
"}},{"node":{"frontmatter":{"title":"Automated Newsletter Generator with CrewAI and Exa","tech":["LLM","AI Agents","CrewAI","LangChain","Streamlit"],"github":"https://github.com/oelbourki/Automated-Newsletter-Generator-with-CrewAI-and-Exa","external":"https://github.com/oelbourki/Automated-Newsletter-Generator-with-CrewAI-and-Exa","order":3,"date":"2024-01-01"},"html":"Developed an AI-driven system using CrewAI to manage research agents for news fetching and summarization. Integrated Exa for data aggregation and employed LangChain and ChatGPT for generating concise summaries. Designed a Streamlit GUI for easy interaction and automation of the newsletter process.
"}},{"node":{"frontmatter":{"title":"End to End Kidney Disease Classification System MLops","tech":["Mlflow","UI","DVC","AWS","Mlops"],"github":"https://github.com/oelbourki/End-to-End-Kidney-Disease-Classification-System-MLops","external":"https://github.com/oelbourki/End-to-End-Kidney-Disease-Classification-System-MLops","order":4,"date":"2024-01-01"},"html":"Led the end-to-end implementation of a kidney disease classification project, utilizing modern software development methodologies and tools including GitHub, MLflow, DVC, Dockerization, and AWS CI/CD for seamless deployment and efficient workflow management.
"}},{"node":{"frontmatter":{"title":"Brain MRI Segmentation with U-Net","tech":["CNN","MRI","Segmentation","IOU","Dice Coefficient"],"github":"https://github.com/otmane-elbourki/brain-mri-segmentation","external":"https://brain-mri-segmentation.com","order":5,"date":"2021-05-18"},"html":"Designed and implemented a U-Net model for brain MRI segmentation with IOU of ~0.99 and Dice Coefficient of ~0.80, leveraging data augmentation and normalization techniques to optimize model performance.
"}},{"node":{"frontmatter":{"title":"matrix","tech":["Python","CUDA","Linear Algebra","NumPy","High-Performance","AI"],"github":"https://github.com/oelbourki/matrix","external":"","order":6,"date":"2024-03-01"},"html":"High-performance Python & CUDA library for vector and matrix operations, optimized for AI and numerical computing.
"}},{"node":{"frontmatter":{"title":"Inception of Things","tech":["Kubernetes","K3s","K3D","ArgoCD","GitOps","GitLab","CI/CD"],"github":"https://github.com/oelbourki/Inception-of-Things","external":"","order":7,"date":"2024-06-01"},"html":"Kubernetes & GitOps learning project with K3s/K3D clusters, ArgoCD workflows, and GitLab integration for CI/CD practice.
"}}]}}}