Generative AI (GenAI) for DevOps and DevSecOps Engineers

Explore the fundamentals of generative AI, its real-world examples, and how machine learning powers models like GPT, DALL-E, and Codex for organizations.

Explore the basics of generative AI, including code and image generation, text production, and devsecops-focused use in CI/CD pipelines with tools like Copilot, ChatGPT, and Dali.

Explore generative ai basics by defining artificial intelligence, machine learning, and the three learning types—supervised, unsupervised, and reinforcement—along with algorithms and models.

Explore what a model is in lay terms using a cake-baking analogy, covering data collection, training, refinements, and deploying a model for tasks such as text, image, and video generation.

Explore tokens and embeddings as fundamental artificial intelligence concepts, and learn how prompts are tokenized—word, subword, or character level—then converted to embeddings for neural networks.

Master prompt engineering for DevOps and DevSecOps by applying zero-shot, one-shot, few-shot, and chain-of-thought strategies to create effective prompts that guide AI models like ChatGPT, Copilot, and DALL-E.

Explore how ChatGPT powers a large language model, the Generative Pre-trained Transformer (GPT), through unsupervised, supervised, and reinforcement learning from human feedback on a vast Common Crawl dataset.

Explore how Delhi generates images from text using a transformer-based neural network trained on a vast text-image pair dataset from the internet, with unsupervised learning and supervised fine-tuning on GPT-3.

Explore how GitHub Copilot works by detailing its Codex model, transformer architecture, and the unsupervised and supervised learning processes that train it on vast public code.

Generate a declarative Jenkins file for a DevOps workflow using generative AI, and learn to craft accurate prompts while integrating docker, AWS ECR, and Kubernetes deployment.

Learn to complete DevSecOps tasks with GenAI by enhancing a Jenkins pipeline with security: SAST with Sonar Cube and DAST with Zap.

Acquire prerequisites to run generative AI offline by downloading local models and installing Python and PyCharm to send prompts to offline AI for DevOps and DevSecOps.

Learn to run large language models offline on your local system with GPT for all and Hugging Face, using llama three, then generate a deployment YAML for ECS or EKS.

Explore a hands-on generative ai case study for devops and devsecops. Use embeddings and a private gpt to identify errors in a rest api log and outline fixes.

Install Microsoft C++ build tools on Windows, using the Visual Studio installer to select C++ version 143 64-bit and Windows 11 SDK, enabling private GPT on your system.

Install Ulama to run local LLMs on Windows and enable private GPT offline; understand RAM requirements and model sizes to run llama 3, mistral, code llama on your hardware.

Install miniconda on Windows to provide a lightweight isolated Python environment, verify conda, and learn commands for activating, creating, and cleaning environments.

Clone the personal GPT repo on a Windows system to read incident logs, identify errors, and propose mitigations, using a conda environment, embeddings, and a local AI model.

Clone the Gen I personal GPT repo, set up a conda environment with Python 3.11, activate it, and install required packages to run personal GPT locally to identify log errors.

Install dependencies, pull the mistral model, ingest documents to create embeddings in a vector store, then run personal gpt locally to identify 500 errors in the log and suggest fixes.

Clone the Spring Boot weather inquiry repo to fetch live weather data via a weather API, feed it to the Ulama/Mistral model, and explore the weather data service.

Clone the Ulama spring boot repo locally, configure a weather function with API key and URL, and expose a city weather endpoint feeding live data to Ulama models with logging.

Conduct a code walkthrough to fetch and ingest live weather data into an AI model such as Mistral via Llama, and validate results through API calls.

Install Docker Desktop on Windows, download and run the installer, and sign in to Docker Hub; then enable Virtual Machine Platform and upgrade to WSL 2 to run Docker smoothly.

Learn to run Open Web UI on your local system, work with multiple llama and OpenAI models offline, upload code, and get model explanations.