Pragmatic AI Labs Courses

Deterministic LLM programming

Build production LLM systems on AWS Bedrock: configure knowledge bases with S3 for RAG, compile llama.cpp for local GGUF inference, and evaluate LLM quality.

• Coursera
• 4 weeks, 1 hour a week
• Self-Paced
• Paid Course

Enterprise AIOps with Amazon Q Business

Deploy Amazon Q Business for enterprise knowledge access, script CloudShell for Bedrock APIs, control AI costs with anomaly detection, and build RAG workflows on S3 knowledge bases.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

LLM Security and Vulnerabilities

Identify and defend against LLM vulnerabilities such as prompt injection, insecure output handling, model theft and excessive agency, applying sanitization, output validation and rate limiting in a security audit.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

Multi-modal AI

Turn screenshots into production code: prompt with visual context, code with GitHub Copilot in VS Code, and validate builds using Playwright and MCP.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

Privacy-Conscious Development with AI Assistants

Use AI coding assistants without exposing sensitive data: run AI-assisted security reviews with Claude Code and automate scanning with GitHub Advanced Security, Grype, and Dependabot.

• Coursera
• 3 weeks, 2 hours a week
• Self-Paced
• Paid Course

Prompt Architecture and NLP on Amazon Bedrock

Design versioned prompt architectures on Amazon Bedrock and build NLP pipelines with chain-of-thought agents, AWS CLI invocation, and custom Rust tools like SVGen.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI Code Review Automation with GitHub Actions

Build an AI-powered code review bot as a GitHub Action: define criteria with pmat, craft review prompts, test locally, deploy on pull requests, publish to GitHub Marketplace.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI Debugging and Test-Driven fixes

Debug systematically with AI: isolate bugs with failing tests, feed tools logs, traces and project docs for context, and hunt defects proactively without regressions.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI Orchestration: From local models to cloud

Orchestrate AI across local and cloud: run Ollama Modelfiles, llamafile and Rust Candle with CUDA, monitor GPUs, and deploy on AWS Spot instances.

• Coursera
• 4 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI Security and Governance on AWS

Design AI security architectures on AWS: configure Bedrock guardrails with content filters and PII detection, audit Bedrock API calls with CloudTrail, and apply SageMaker Clarify bias detection.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI Tooling Capstone: Serverless Multi-Model Systems

Build a serverless multi-model AI system on AWS: route requests across Amazon Bedrock and Ollama, deploy Rust functions with Cargo Lambda, and configure YAML-driven prompts.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AI-Powered Data Pipelines with Deno

Build AI-powered data pipelines with Deno: configure tasks in deno.json, compose task playbooks, and use deno compile, deno doc, and deno vendor for production builds.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

AWS Generative AI and Foundation Models

Build generative AI solutions on AWS with Amazon Bedrock, RAG pipelines, and Amazon Q Developer, then compile llama.cpp and deploy GGUF-packaged open-source LLMs on EC2 GPUs.

• Coursera
• 2 weeks, 3 hours a week
• Self-Paced
• Paid Course

AWS Intelligent Applications with Amazon Bedrock

Build intelligent applications on Amazon Bedrock: compare foundation models in the console, write clients in Bash curl and Rust, and construct knowledge-base-backed agents.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course

Conversational Bot Architecture with Rust and Deno

Build multi-platform conversational bots in Rust and Deno: design Cargo workspaces, run Tokio async event loops, connect Amazon Bedrock with Claude, and deploy a Discord bot.

• Coursera
• 3 weeks, 1 hour a week
• Self-Paced
• Paid Course