Learn by Source
Udemy

AI Engineer Core Track: LLM Engineering, RAG, QLoRA, Agents

4.7(39,752) on Udemy·327K enrolled
Intermediate 33.5 hours English Course Certificate
SkillsLLM engineeringRAGFine-tuningQLoRAAI agentsPython

Is this course right for you?

Our take
An intensive, build-along bootcamp for becoming a hands-on LLM engineer, aimed at intermediate developers. You ship eight real projects.

Good for: python developers who want to build real LLM applications and agents.

Skip if: you are new to Python or machine learning, or you only want to use AI tools.

Across the eight weeks you build a web-scraping generator, a multimodal support agent, a RAG knowledge worker, a QLoRA-fine-tuned open model, and an autonomous multi-agent deal-finder deployed serverless — experimenting with 20-plus frontier and open models. Where shorter courses teach single techniques, this goes end to end and unusually deep on fine-tuning and deployment.

It assumes Python and some ML, so it is not a first course — start with an intro if you are new. It needs a small (~$5) budget for frontier-model API calls, though open-source models can substitute. Donner refreshed it in 2026 for the latest models; on price, wait for a Udemy sale.

Comparison · LBS

Compare alternatives for AI Engineer Core Track: LLM Engineering, RAG, QLoRA, Agents

Same topic, different options. We surface the trade-offs others hide so you can pick the course that actually fits your time, budget, and goals.
Udemy4.7(39,752)
AI Engineer Core Track: LLM Engineering, RAG, QLoRA, Agents
Price
Paid
Paid, frequently discounted · lifetime access (+ small API costs)
Duration
33.5 hrs
Level
Intermediate
Certificate
Course Certificate
DataCamp
Designing Agentic Systems with LangChain
Price
Paid
DataCamp subscription
Duration
3 hrs
Level
Intermediate
Certificate
Course Certificate
View course →
DataCamp
Retrieval Augmented Generation (RAG) with LangChain
Price
Paid
DataCamp subscription
Duration
3 hrs
Level
Intermediate
Certificate
Course Certificate
View course →
DataCamp
Introduction to LLMs in Python
Price
Paid
DataCamp subscription
Duration
3 hrs
Level
Intermediate
Certificate
Course Certificate
View course →
Prices & availability can change — confirm on the provider's site. We're not affiliated with any single provider.

About this course

AI Engineer Core Track: LLM Engineering, RAG, QLoRA, Agents is an intensive, eight-week build-along bootcamp for becoming a hands-on LLM engineer. Across about 33.5 hours you ship eight real projects — from a web-scraping brochure generator and a multimodal support agent to a knowledge worker built on RAG, a fine-tuned open-source model (QLoRA), and an autonomous multi-agent deal-finder deployed serverless on Modal — while experimenting with 20+ frontier and open-source models.

Browse all AI & ML courses Browse all Udemy courses

Instructor

ED
Ed Donner
Udemy instructor

Taught by Ed Donner, an entrepreneur and AI leader with 20+ years of experience who has co-founded and sold an AI startup and led technology teams at major financial institutions.

Frequently asked questions

Comfortable Python is essential — this is a hands-on engineering course, not an overview. You should be able to read and write real code, work with APIs, and use the command line without hand-holding. You do not need prior machine-learning theory or maths, since the focus is building with existing models rather than training from scratch, but arriving without solid Python will make the projects a struggle.

The four pillars of modern AI engineering: working with large language models, retrieval-augmented generation (RAG), fine-tuning with QLoRA, and building agents. It is structured around eight real projects — a brochure generator that scrapes websites, a multi-modal customer-support agent with function-calling, a meeting-minutes tool from audio, a Python-to-C++ optimiser, and a RAG knowledge-worker among them — so you build shippable things rather than just watching.

Real production tooling rather than toys. You work with frontier-model APIs alongside open frameworks — LangChain for orchestration, Chroma for vector storage, Hugging Face for open models, Gradio for quick UIs, plus Weights & Biases and Modal for training and deployment. Experimenting across twenty-plus models is part of the point, so you leave knowing how to pick and combine tools, not just one stack.

Yes, and it is worth planning for. Calling frontier-model APIs and running fine-tuning jobs on cloud services cost money based on usage — usually modest for coursework, often just a few dollars, but not zero. You can keep it low by using lower-cost models and free tiers where the course allows, but budget a little for API and compute credits on top of the course price.

This is one of the more current LLM-engineering courses available, kept aligned with the fast-moving tooling and highly rated across a large enrolment. For a Python developer wanting to move into AI engineering with genuinely practical, build-heavy projects, it is a strong choice. The field shifts quickly, so expect to keep learning after it — but as a structured, hands-on foundation it holds up well.
Paid
Paid, frequently discounted · lifetime access (+ small API costs)
Enroll now