Home / Certificates / CPP-35
Professional Certificate · Master's level · CPP-35
Applications Built on Large Language Models
For: Developers, engineers and digital project managers who design assistants, document search engines or other generative applications.
Learning outcomes
- Design an application built on a large language model, choosing the adaptation method.
- Evaluate the quality, faithfulness to sources and safety of a generative model's outputs.
- Justify the choice between a locally hosted open model and an external model service.
Programme over 4 weeks
| Week | Content |
|---|---|
| Week 1 | Large language models: principles, capabilities and limitations; choosing between a locally hosted open model and an external service in terms of cost, sovereignty and the data processed. |
| Week 2 | Adaptation: prompt engineering, retrieval-augmented generation over a document base; criteria for choosing the method, in a virtual lab. |
| Week 3 | Evaluation: test sets, human evaluation, faithfulness to sources, hallucinations, bias; evaluation in French and in national languages. |
| Week 4 | Professional simulation: retrieval-augmented generation application over a document base provided, with its evaluation protocol. |
Assessment
Marked professional simulation in a virtual lab, assessed against a rubric (70%): retrieval-augmented generation application and evaluation protocol; online knowledge-check quiz (30%). The document base is provided by the instructor and contains no personal data.
The certificate is awarded with a mark of at least 10 out of 20.
Optional: Fine-Tuning and Rights in Generated Content
An additional 1-credit module over 2 weeks (CPP-35C), awarded as a separate certificate; together with this certificate, it makes up the full course component of the original programme.
- Parameter-efficient fine-tuning; diffusion models and multimodal models: principles, capabilities and limitations.
- Intellectual property in generated content and training data under loi n°2008-09 (Copyright and Related Rights Act); transparency towards users.
At your own pace: the self-paced pathway
In a class, this certificate takes 4 weeks, on the cohort's dates. On the self-paced pathway, you start whenever you like, online, and progress at your own pace. The price is the same.
Estimate how long my pathway will take
- Duration
- about 4 weeks
- Estimated end
- 4 weeks after you start
- Examination
- Online, subject to available slots, no earlier than 4 weeks after you start
- Payment
- In a single payment
An indicative estimate, for a start today and 40 hours of work. You can change your pace along the way; the pathway must be completed within 12 weeks (about 3 months) of your start. Your e-Campus access opens with your first payment.
Admission
Programming experience and basic knowledge of machine learning (Bachelor's degree level in computer science or equivalent experience assessed on application). Admission based on application; detailed requirements set by the school.
The credits you earn are recognised without further assessment if you later join the programme they come from, Professional Master's degree in Artificial Intelligence, within a period set by the certificate regulations.