Large Language Model Engineering (LLM Engineering) refers to the emerging best-practices and tools for pretraining, post-training, and optimizing LLMs prior to production deployment. Pre- and post-training techniques include unsupervised pretraining, supervised fine-tuning, alignment, model merging, distillation, quantization. and others.

Engineering toolkit for building scalable AI systems, including model training pipelines, LLM optimization utilities, and data engineering tools.

A ground-up LLM engineering project: tokenizer → architecture → training → scaling laws → inference. Starts at 80M, engineered to scale into 1B+ models with minimal changes. Clean, research-ready code for anyone serious about understanding and building LLMs from first principles.

The world's no 1 classroom for incubating AI talent. 4 months of hands-on, no-BS technical training in ML, LLMs and Production AI engineering.

GeneticPromptLab uses genetic algorithms for automated prompt engineering (for LLMs), enhancing quality and diversity through iterative selection, crossover, and mutation, while efficiently exploring minimal yet diverse samples from the training set.

Web2LLM.txt – A fast, open-source website-to-LLM context file generator. Paste any https:// URL and instantly get a clean llm.txt file with token & cost estimation—ideal for RAG, prompt engineering, and AI training workflows.

An LLM-powered automated tutoring program that will converse with you on any given branch of topics (technical or soft). Has practical uses for Quantitative training in Finance, Economics training in Investment Banking, Software Engineering, or Data Science. Can teach basic STEM topics as well as the arts and humanities.

An engineering-focused project covering data acquisition, dataset curation, tokenization, training, and evaluation of an AI/ML-focused LLM.

Course on LLMs and Prompt Engineering. Covers LLM fundamentals, training, evaluation, prompting techniques, RAG, multimodal capabilities, agents, MCP, and LLM-powered software engineering tools.

The Full-Stack LLM Engineering Playbook. Architectural patterns for Agents (MCP) & RAG, coupled with advanced Post-Training recipes (SFT, DPO, QLoRA) for domain adaptation. Covers Data Pipelines, Evaluation Frameworks, and System Design.

This project focuses on text-generation LLMs, offering a deep dive into building and fine-tuning large language models for generating human-like text. It covers key techniques in training, prompt engineering, and model optimization, enabling the creation of powerful, context-aware text generation applications for diverse use cases.

LLaMA Factory is an end-to-end LLM fine-tuning and deployment pipeline, integrating data augmentation & engineering, cloud-based training, and cloud data management. With one-click fine-tuning and local deployment, it enables rapid iteration of models while keeping your infrastructure flexible and secure. Perfect for enterprise-grade applications.

This is designed for generating prompts that train LLMs to reason, analyze, and generate prompts—for internal tooling, agents, or fine-tuning.

🚀 Production ML Engineering: The Complete Guide to Distributed LLM Training & Serving Master the art of building, optimizing, and deploying large-scale ML systems in production environments 🎯 This repository is your complete handbook for becoming a production LLM machine engineer.

Training materials for LLM engineering

Training LLMs, Jupyter Notebook, ChatGPT Prompt Engineering for Devs

All things AI Engineering : Models, Transformers, LLMs, FrontEnd, BackEnd, Distributed Training, On Cloud

NLP research project: automated prompt engineering method for training LLM on logic and reasoning

Production-grade LLM fine-tuning tutorial: Dataset engineering, LoRA training, vLLM serving - completely self-hosted

Week 4 of LLM Engineering Certification: Learn memory limits, distributed training, and production-ready workflows.

Benchmarking AdamW, SophiaG and AdamSNSM optimizers for training NanoGPT under resource‑constrained conditions (single GPU with 8GB VRAM). Includes reproducible training pipeline, experimental results and engineering practices for memory‑efficient LLM training.

An end-to-end Data Science pipeline for analyzing mathematical misconceptions, featuring automated ETL, MySQL integration, and feature engineering for LLM training. Dockerized & CI/CD enabled.

Interactive Gen AI training site built for covering LLM fundamentals, RAG architecture, Copilot 365, prompt engineering, and a role-specific prompt library. Pure HTML/CSS/JS.

Final project of the course "Large Scale AI Engineering" at ETH Zürich, FS2025. Implementation and benchmarking of pretokenization and Distributed Data Parallel (DDP) for efficient LLM training on the CSCS Alps supercomputer.

DocuShield is a hybrid AI document risk analysis system that processes legal contracts using AWS Textract for OCR and Groq-powered LLMs for reasoning. It runs on Kubernetes, stores results in DynamoDB, supports clause-level RAG, and focuses on real-world AI system engineering, not model training.

No description available

No description available

Learning LLM engineering through an online course on Udemy taught by Ed Donner.

No description available

No description available