Welcome to this cutting-edge AI Engineering project! This repository showcases a comprehensive exploration of various AI technologies, software engineering practices, and advanced programming concepts. It's structured to demonstrate a progressive learning journey through different aspects of AI and software development and is designed to follow my learnings at the AI Makerspace "LM-Engineering-Foundations-to-SLMs" bootcamp.
- Week 1: Foundations and LLM Studies
- Week 2: Advanced Concepts and Implementations
- Week 3: AI Agents, CUDA, and DevOps
This week focuses on establishing core concepts and diving into Large Language Model (LLM) studies.
- hoc.py: Implementation of higher-order functions in Python.
- Latin Side Quest: A fascinating exploration of advanced NLP techniques applied to classical Latin:
- DiachronicWordEmbeddings.py
- Macron-AwareSubwordTokenization.py
- MacronPositionalEncodingImproved.py
- MacronPositionEncoding.py
- Prosody-AwareAttentionMechanism.py
A series of in-depth studies on Large Language Models:
- LLM_Studies-Nov-17-2024.md
- LLM_Studies-Nov-18-2024.md
- LLM_Studies-Nov-19-2024.md
- LLM_Studies-Nov-21-2024.md
- LLM_studies-Nov-22-2024.md
These studies cover various aspects of LLMs, including architecture, training methodologies, and applications.
Week 2 delves deeper into AI and software engineering concepts, featuring a mix of theoretical studies and practical implementations.
- AfterKandV.md: Post-analysis of key AI concepts.
- AutoencodingNotes.md: Comprehensive notes on autoencoding techniques.
- CreatingDeepandWideNeuralNetworks.md: Guide to architecting complex neural networks.
- Positional_Encoding_Formula.md: Detailed explanation of positional encoding in transformer models.
- almost_orthogonal.py: Python script demonstrating almost orthogonal vector generation.
- llama_tests.py: Test suite for LLaMA model implementations.
- yi.py: Implementation related to the Yi language model.
- HardcoreKubernetesk3d.md: Advanced Kubernetes configurations using k3d.
- k3d-secure.yaml: Secure configuration for k3d clusters.
-
llamagguf-example: Rust implementation of LLaMA model quantization:
- Cargo.lock
- Cargo.toml
- src/main.rs
-
movies_demo_ml_python: Movie recommendation system using machine learning:
- corpus_prep.py
- corpus.txt
- tokenizer/vocab.txt
-
nvidia: NVIDIA-specific AI model implementations:
- Dockerfile
- model.py
- Research.md: Compilation of research findings and insights.
- RustLifetimes.md: Deep dive into Rust's lifetime concept.
- RustTraits.md: Comprehensive guide to Rust traits.
The final week focuses on advanced AI applications, GPU computing, and DevOps practices.
Three innovative chatbot implementations in Rust:
- comedian_chatbot:
- rig_chatbot:
- self_improving_perplexity_chatbot:
- check-nvidia-cuda.sh: Script to verify NVIDIA CUDA installation.
- cuda_program.cu: CUDA C program demonstrating GPU computation.
- cuda.py: Python interface for CUDA operations.
- InstallingCudaisBig.md: Comprehensive guide to CUDA installation.
- Rust_cuda_research.md: Research findings on integrating Rust with CUDA.
- k3d-secure.yaml: Enhanced secure configuration for k3d.
- llm-cli.md: Command-line interface guide for LLM operations.
- Modelfile: Configuration file for model deployment.
- UseEncryptionwithJenkins.md: Guide to implementing encryption in Jenkins pipelines.
Zarf-related configurations for AI deployment:
- Jenkinsfile
- WhyUseZarfforAI.md
- zarf-podinfo.yaml
- ZarfVLLM.yaml
This project demonstrates a comprehensive approach to AI engineering, covering everything from foundational concepts to cutting-edge implementations. It showcases proficiency in various programming languages, DevOps practices, and AI technologies, making it an impressive portfolio piece for AI engineering students and professionals.