What you'll learn
- Deploy SaaS LLM apps to production on Vercel, AWS, Azure, and GCP, using Clerk
- Design cloud architectures with Lambda, S3, CloudFront, SQS, Route 53, App Runner and API Gateway
- Integrate with Amazon Bedrock and SageMaker, and build with GPT-5, Claude 4, OSS, AWS Nova and HuggingFace
- Rollout to Dev, Test and Prod automatically with Terraform and ship continuously via GitHub Actions
- Deliver enterprise-grade AI solutions that are scalable, secure, monitored, explainable, observable, and controlled with guardrails.
- Create Multi-Agent systems and Agentic Loops with Amazon Bedrock AgentCore and Stands Agents
Requirements
- While it’s ideal if you can code in Python and have some experience working with LLMs, this course is designed for a very wide audience, regardless of background. I’ve included a whole folder of self-study labs that cover foundational technical and programming skills. If you’re new to coding, there’s only one requirement: plenty of patience!
- The course runs best if you have a small budget for APIs and Cloud Providers of a few dollars. But we monitor expenses at every point, and it's always a personal choice.
Description
This is the course that more of my students have asked for than any other course — put together.
One student called it:
“The missing course in AI.”
This course is for:
Entrepreneurs
Enterprise engineers
…and everyone in between.
It’s not just about RAG — although we’ll work with RAG.
It’s not just about Agents — but there will be many Agents.
It’s not just about MCP — but yes, there will be plenty of MCP too.
This course is about:
RAG, Agents, MCP, and so much more… deployed to production.
Live.
Enterprise-grade.
Scalable, resilient, secure, monitored — and explained.
You’ll ship real-world, production-grade AI with LLMs and agents across Vercel, AWS, GCP, and Azure, going deepest on AWS.
Across four weeks you’ll take four products to production:
Week 1
You’ll launch a Next.js SaaS product on Vercel and AWS,
with AWS App Runner and Clerk for user management and subscriptions.
Week 2
You’ll become an AI platform engineer on AWS,
deploying serverless infrastructure using:
Lambda, Bedrock, API Gateway, S3, CloudFront, Route 53
Write Infrastructure as Code with Terraform
Set up CI/CD pipelines with GitHub Actions
— for hands-free deployments and one-click promotions.
Week 3
You’ll gain broad industry skills for GenAI in production:
Deploy a Cyber Security Analyst agent with MCP to Azure & GCP
Stand up SageMaker inference
Build data ingest to S3 vectors
Deploy a Researcher Agent using OpenAI OSS models on Bedrock + MCP
Week 4
You’ll go fully agentic in production:
Architect multi-agent systems with:
Aurora Serverless, Lambda, SQS
JWT-authenticated CloudFront frontends
LangFuse observability
Overview of AWS Agent Core
By the end, you’ll know how to:
Pick the right architecture
Lock down security
Monitor costs
Deliver continuous updates
Everything needed to run scalable, reliable AI apps in production.
Course sections (Weeks & Projects)
Week 1
SaaS App Live in Production with Vercel, AWS, Next.js, Clerk, App Runner
Project: SaaS Healthcare App
Week 2
AI Platform Engineering on AWS with Bedrock, Lambda, API Gateway, Terraform, CI/CD
Project: Digital Twin Mk II
Week 3
Gen AI in Production with Azure, GCP, AWS SageMaker, S3 Vectors, MCP
Project: Cybersecurity Analyst
Week 4
Agentic AI in Production: Build and deploy a Multi-Agent System on AWS (Aurora Serverless, Lambda, SQS),
with LangFuse and Bedrock AgentCore
Capstone Project: SaaS Financial Planner
Who this course is for:
- If you're excited about the idea of deploying Gen AI and Agents live in production - then this course is for you.
Instructors
Join 3.8M+ learners who study with Ligency.
With a 4.6 instructor rating, >1.1 M reviews, and 126 courses in 12 languages, we help engineers, leaders, and teams master the skills that power today’s AI revolution - then ship real results.
We start where the real world starts: with large language models and the products they power. You’ll learn the foundations of AI and Generative AI (gen AI), then ship production-grade systems - chatbots, copilots, automations, and AI agents. We go deep on LLM engineering: retrieval (RAG), evaluation, observability, safety, and the patterns teams use to run agentic systems at scale.
Our stack is practical and current. You’ll prototype fast with Python, LangChain, and LangGraph; explore models from OpenAI, Gemini, and Claude (including Claude Code); fine-tune and serve with Hugging Face and Ollama; and take it to production on AWS - from Bedrock to event-driven services. Need automation? We wire it together with n8n, clean interfaces, and CI/CD. Along the way you’ll master prompt engineering that holds up under load.
Where this leads: roles that ship. AI Engineer and LLM Engineer for those who love building; platform and MLOps paths for those drawn to reliability at scale; product and leadership tracks for the people moving Agentic AI from slide decks to business outcomes. The through-line is the same: learn fast, build faster, measure everything, iterate.
Start with our best selling course:
AI Coder: Complete Claude Code & Coding Agents Course - build complete products at speed with AI coding agents like Claude Code, Cursor, Copilot and Codex, no coding background required.
AI Engineer Core Track: LLM Engineering, RAG, QLoRA, Agents - a hands-on path from your first prompt to production patterns: 20+ models, RAG, QLoRA fine-tuning, and agents with LangChain/LangGraph.
AI Engineer Agentic Track: The Complete Agent & MCP Course - design, orchestrate, and deploy robust AI agents across OpenAI Agents SDK, CrewAI, LangGraph and MCP.
AI Engineer Production Track: Deploy LLMs & Agents at Scale - scaling patterns for pipelines, monitoring, and enterprise rollout on AWS with Bedrock, Google Cloud Platform, Azure and MLOps.
AI Builder: Create Agents, Voice Agents & Automations in n8n - wire up low-code AI agents, voice agents and business automations in n8n with ElevenLabs, RAG and MCP.
Practical Guide to AI Agents & Agentic AI with Claude Cowork - build six working no-code AI agents on your real tools (Gmail, Slack, Notion, Calendar) with Claude Cowork and MCP.
AI Leader: Generative AI & Agentic AI for Leaders & Founders - a concise playbook for strategy, governance, and ROI with Generative AI and agents.
If your goal is to level up quickly and ship something real, join us. Learn the concepts, touch the tools, build the thing - then take it to users. That’s the Ligency way.
Ed Donner is a technology leader and repeat founder of AI startups. He’s the co-founder and CTO of Nebula, the platform to source, understand, engage and manage talent, using Generative AI and other forms of machine learning. Nebula’s long-term goal is to help people discover their potential and pursue their reason for being.
Previously, Ed was the founder and CEO of AI startup untapt, an Accenture Fintech Innovation Lab company, acquired in 2021. Before that, Ed was a Managing Director at JPMorgan Chase, leading a team of 300 software engineers in Risk Technology across 3 continents, after a 15-year technology career on Wall Street. Ed holds a patent for a Deep Learning matching engine issued in 2023, and an MA in Physics from Oxford.