AI in the Software Development Process – Code Governance, Testing and Safe Use of Artificial Intelligence

Training process

Training needs analysis

If you have specific requirements regarding the training programme, we will carry out a training needs analysis for you. This will guide us on which aspects of the programme should receive greater emphasis, so that the training programme meets your specific needs.

What will you gain?

Practical AI Across SDLC - You will learn how to use AI tools deliberately across analysis, coding, testing, and deployment, so you can move faster while still keeping quality and engineering decisions under control.

AI Code Governance Rules - You will learn how to define policies, standards, and oversight for AI-assisted code, so you can introduce these practices in your team without confusion, gaps, or unclear ownership.

Safer AI-Generated Code - You will learn to spot vulnerabilities, risky dependencies, and unsafe model suggestions, helping you reduce security flaws before changes are merged into your repository or release flow.

Stronger Output Validation - You will master ways to detect hallucinations, logic flaws, and misleading AI recommendations, so you can judge more accurately when a model response is reliable enough for real use.

Testing AI and RAG Systems - You will learn how to design tests for LLM and RAG-based applications, assess response quality, and verify the reliability of integrations with APIs, databases, and business applications.

AI in CI/CD Automation - You will see how to use AI for test generation, documentation, refactoring, and quality reporting, and how to add these activities safely to CI/CD pipelines and delivery workflows.

Compliance and Licensing - You will understand how to account for the AI Act, GDPR, copyright, and open source licenses in daily work, reducing legal and compliance risk when using AI-supported code and services.

Team Standards Implementation - You will be ready to define roles, checklists, acceptance criteria, and audit practices for AI usage, helping your team work in a more consistent, measurable, and controlled way.

Training programme

1. Introduction to AI in the software development process

the role of artificial intelligence in modern Software Development,
the use of AI at individual stages of the software life cycle (SDLC),
AI-assisted Development – possibilities and limitations,
overview of tools supporting programming: GitHub Copilot, ChatGPT, Claude Code, Cursor,
integration of AI tools with development environments,
benefits and risks resulting from the automation of code creation.

2. AI Code Governance – management of the use of AI in programming

basic principles of AI Code Governance,
creating policies and standards for the use of AI in IT teams,
defining permissible uses of AI tools,
responsibility for code generated by artificial intelligence,
principles of supervision over the process of generating and modifying code,
documenting the use of AI in IT projects,
quality control and compliance of code with the organization's standards,
managing access to AI tools and models,
auditing the use of AI in software development processes.

3. Security of AI-generated code

the most common threats related to AI-generated code,
identification of security vulnerabilities in code,
detection of unsafe libraries and dependencies,
the risk of generating code inconsistent with good security practices,
protection of source code and confidential information,
secure transmission of data and code fragments to AI models,
management of secrets, tokens and API keys,
use of OWASP standards in security assessment,
analysis of the risk associated with the use of external AI models.

4. Testing and verification of code generated by AI

methods of assessing the correctness of code generated by AI,
Code Review supported by artificial intelligence,
detection of logical, syntax and architectural errors,
unit tests (Unit Testing),
integration tests (Integration Testing),
regression tests and quality control automation,
generation of test scenarios and test cases using AI,
verification of test coverage of the code,
analysis of code quality, performance and maintainability,
detection of hallucinations and incorrect suggestions of AI models.

5. Testing applications and systems using AI

specifics of testing applications based on language models (LLM),
testing the quality and correctness of AI model responses,
methods for evaluating the effectiveness of AI solutions (AI Evaluation),
testing the reliability and repeatability of model operation,
verification of the operation of RAG systems,
testing the integration of AI models with applications and databases,
detecting hallucinations and data interpretation errors,
resilience testing against Prompt Injection,
testing the security of API interfaces,
monitoring the quality of AI solution performance after deployment.

6. Automation of software development processes using AI

use of AI for code generation and refactoring,
automation of technical requirements analysis,
generation of technical documentation,
support for detecting and fixing errors (Debugging),
automation of test creation and quality reports,
use of AI in Continuous Integration and Continuous Deployment (CI/CD) processes,
integration of security analysis with the software deployment process,
use of AI in managing technical debt,
control of changes in code generated and modified by AI.

7. Responsible use of AI – regulations, compliance and ethics

basics of Responsible AI in IT projects,
AI Act requirements in the context of creating and implementing AI systems,
protection of personal data and compliance with GDPR,
copyright and intellectual property of code generated by AI,
rules for using open source components,
risk of software license infringement,
transparency and the ability to explain the operation of AI solutions,
human oversight over decisions and results generated by AI,
documenting compliance and managing regulatory risk.

8. AI Security – threats and securing AI-based systems

specifics of cybersecurity of systems using AI,
Prompt Injection and manipulation of model instructions,
Data Leakage – the risk of unauthorized disclosure of data,
Model Poisoning and threats related to training data,
unauthorized access to AI tools and agents,
securing integrations with external AI services,
managing permissions and limiting access to resources,
monitoring security incidents,
procedures for responding to AI-related threats.

9. Implementing AI Governance standards in technology teams

developing internal rules for the use of AI,
defining roles and responsibilities in development teams,
creating procedures for verifying code generated by AI,
introducing AI Code Review standards,
establishing acceptance criteria for AI-assisted code,
building security and quality checklists,
integrating AI Governance principles with existing DevOps processes,
monitoring compliance with internal standards,
measuring the quality and effectiveness of AI use.

10. Good practices and development directions of AI in Software Engineering

Human-in-the-Loop model in the programming process,
limiting the risk of excessive reliance on AI-generated code,
maintaining control over software architecture and quality,
safe deployment of AI agents performing programming tasks,
management of permissions of autonomous programming agents,
verification of the results of AI agents' work,
scaling the use of AI in development teams,
developing teams' competencies in safe AI-assisted Development,
the most important recommendations regarding the implementation of AI Code Governance in the organization.

What are the prerequisites for participating in the training?

Programming Basics - You should be comfortable reading and writing code in at least one programming language, because the training focuses on hands-on work with code, tests, and developer tools.

Software Delivery Process - You should understand the main stages of the software lifecycle, such as analysis, implementation, testing, and deployment, so you can map AI topics to real engineering work.

Testing and Quality Basics - You should know the core ideas behind unit tests, integration tests, and code review, because the training expands these areas in the context of AI-assisted code and workflows.

Application Security Awareness - You should know the basic principles of secure software development, including handling data, dependencies, and secrets, so you can follow AI-related risk topics with confidence.