Blog
Accélérez vos tests grâce à l'IA
Adhésion au métier - Méthodes
Informations générales
Objectifs :
Cette formation vous permettra de :
- d'acquérir une compréhension approfondie de l'IA générative pour les tests
- Apprendre des techniques de requête efficaces
- Découvrir et tester des outils d’IA dédiés aux tests :
- ⇒ Anthropic / Claude 3
- ⇒ GPT 3.5 et GPT 4 (OpenAI)
- ⇒ Mistral Small et Mistral Medium (Mistral)
- ⇒ LLM Workbench, Application RAG et la plateforme Gravity (Smartesting)
- ⇒ CodeLLama (Meta)
- ⇒ Sonar Small et Sonar Medium (Perplexity)
- Au fur et à mesure que les technologies évoluent, cette liste sera mise à jour en fonction des éléments les plus pertinents.
- Identifier les limites et les risques des tests optimisés par l’IA
Profil visé :
- Consultant en tests
- Testeur
- Développeur
- Chef de projet
- Spécialiste en automatisation des tests
- Responsable produit
- Chef de projet
- Ingénieur qualité
- Responsable assurance qualité
Exigences obligatoires :
- Expérience en tests logiciels (au moins à un premier niveau)
Introduction :
1. Generative AI for software testing: Introduction
- Generative AI - The basics
- What generative AI brings to software testing
- Using generative AI for testing, general principles (workshop on using the LLM portal for training)
2. Prompt engineering - Requesting a large language model for testing: how to get good results
- Introduction to Prompt Engineering
- Prompting techniques and best practices
- Use cases with practical exercises in 4 workshops: improving existing test cases, test case design, test automation, analyzing anomaly reports
- Plus: tools to help you create your own prompts
- Debriefing and discussion of the skills implemented
3. Managing the risks of generative AI
-
My AI is wrong: how to detect AI errors / evaluation metrics (workshop)
-
My AI doesn't protect my data: managing this risk, what solutions?
-
My AI is biased: concrete examples, how to avoid it
-
Other AI-related risks
-
Environmental risks (workshop on energy costs)
-
Evolution risks
-
AI regulation (European AI Act)
4. LLM-based test applications
-
LLM-based applications: LLMOps,
-
Workshop on Retrieval Augmented
-
Generation (RAG) - Q/R on a large document corpus, LLM Fine-Tuning
-
Integrating LLM-based functions into
-
testing tools: workshop on integrating generative AI functions into a testing tool
-
Autonomous AI agents for testing with demonstration: Gérald - Virtual Manual
-
Tester
-
Summary and discussion
5. Advanced techniques for using generative AI in software testing
-
LLM-based applications
-
Adapting an AI model to test tasks: fine-tuning
-
AI agent testing: towards autonomous or semi-autonomous testing
6. Synthesis: what we've learned
- I'll start using tomorrow: how do I go about it in practice?
- I choose my generative AI model for testing
Teaching Methods:
- 35% theory
- 75% practive : 9+ Workshops & Exercises
Terms of Evaluation:
None
Training Partner:
Our course is in partnership with avec Smartesting.
Request Enrollment
Submit your request - our team will follow up to confirm your seat.
Upcoming Sessions
All sessions are delivered in EN/FR. Minimum 3 participants required for a session to run.
