Issued by Foundryy · Generative AI
Generative AI Foundations
Understand how large language models work, what they are good and bad at, and how to build useful features with them. You will practise prompt design, retrieval-augmented generation and evaluation, and ship a small assistant that answers questions from your own documents.
- Duration
- 3 months
- Study time
- 92 hours
- Modules
- 7
- Cost
- At enrollment
- Generative AI
- Prompt Engineering
- LLM APIs
- RAG
- AI Evaluation
Curriculum
- 110 h
How LLMs work
Tokens and context, Training versus inference, Hallucination and limits
- 212 h
Prompt design
Instructions and examples, Structured output, Prompt testing
- 314 h
Using LLM APIs
Calling a model from code, Streaming and cost control, Safety and privacy
- 418 h
Retrieval-augmented generation
Embeddings, Chunking and search, Grounded answers with citations
- 510 h
Evaluating AI features
Test sets, Human and automated review, Failure analysis
- 66 h
Responsible use
Bias and misuse, Data handling, Disclosure
- 722 h
Capstone
Scoping, Build, Evaluation report and demo
Who can enroll
Basic Python is needed for the build modules. The first module needs no coding.
To earn the certificate
- Complete every module of the curriculum
- Submit a capstone project: a retrieval-augmented question-answering assistant over a document set, with a written evaluation of where it fails
- Pass a final review by a Foundryy reviewer
- Follow Foundryy’s academic-integrity policy: the work you submit must be your own
Issued by Foundryy after the team has reviewed your capstone against the requirements.