Implementation of the model "Hedgehog" from the paper: "The Hedgehog & the Porcupine: Expressive Linear Attentions with Softmax Mimicry"

A fully offline NLP pipeline for extracting, chunking, embedding, querying, summarizing, and translating research documents using local LLMs. Inspired by the fictional mystery of Dr. X, the system supports multi-format files, local RAG-based Q&A, Arabic translation, and ROUGE-based summarization — all without cloud dependencies.

A fully offline NLP pipeline for extracting, chunking, embedding, querying, summarizing, and translating research documents using local LLMs. Inspired by the fictional mystery of Dr. X, the system supports multi-format files, local RAG-based Q&A, Arabic translation, and ROUGE-based summarization — all without cloud dependencies.

A hands-on course repository for Evaluating AI Agents, created with Arize AI, that teaches you how to systematically evaluate, debug, and improve AI agents using observability tools, structured experiments, and reliable metrics. Learn production-grade techniques to enhance agent performance during development and after deployment.

AI-Augmented ESG Analysis - Oliver led efficiency and local model use fork

The course teaches how to fine-tune LLMs using Group Relative Policy Optimization (GRPO)—a reinforcement learning method that improves model reasoning with minimal data. Learn RFT concepts, reward design, LLM-as-a-judge evaluation, and deploy jobs on the Predibase platform.