Artificial Intelligence in Software Development

Course ID : AI-002

Duration In-class (в days) : 24 academic hours

Duration Online : 24 academic hours

Сurriculum : in-class, Virtual Instructor-Led Training

Overview

The “Artificial Intelligence in Software Development” course focuses on methods for integrating AI components into the software development lifecycle to improve process efficiency.

The course examines the practical application of AI in software development, from code generation and testing to automated reviews and documentation.

The program covers the principles of language models, prompt engineering, API integration, and embedding AI in CI/CD and IDEs. Special attention is given to agent-based systems, RAGs for codebases, and methods for quality control of generated results.

Attendees will learn best practices, tools, and frameworks for AI-powered software development, as well as the risks and privacy issues associated with using AI tools in an enterprise environment.

Audience for this course

  • Technical Managers
  • Software Architects
  • IT Landscape Architects
  • Analysts
  • Software Developers

Prerequisites for this course

  • Basic knowledge of the Python programming language.
  • Understanding of the software development lifecycle and testing fundamentals.
  • Knowledge of typical software architecture patterns.
  • Experience with AI tools is not required—all necessary approaches and methods are covered during training.

Outcomes

Upon completion of this course, students will gain the skills to:

  • Apply AI and AI use cases in software development processes.
  • Analyze AI applicability problems and formulate them in terms of prompts and agent interactions.
  • Apply various types of LLM (text-based, multimodal) and AI tools (chatbots, IDE plugins, agent systems) to solve SDLC problems.
  • Use prompt engineering techniques: system instructions, role-based prompts, and “generation → review → fix” cycles.
  • Integrate AI via APIs, build chain pipelines, and manage sessions.
  • Work with RAG on documents and the codebase, using embeddings and semantic search.
  • Integrate AI into the development pipeline: CI/CD, IDEs (Copilot, Continue, GigaCode), and code agents (Cursor, Claude Code).
  • Understand AI implementation methodology, including MLOps pipelines, A/B testing, and monitoring.
  • Assess AI performance using metrics and mitigate agent error.
  • Understand AI limitations (hallucinations, context drift, inconsistency) and how to mitigate them.

Outline

1. The history of software development assistants.
2. An overview of possible AI use cases in software development, and the place of AI in the software development pipeline.
3. Collaborative case study analysis to analyze potential AI application points.
4. Model operating principles. Required data for model operation.
5. Prompt engineering for developers. API usage.
6. Prompt engineering practice for developers.
7. An overview of popular AI frameworks, libraries, platforms, and cloud services.
8. Using AI for test writing.
9. Integrating AI into developer tools (IDE, CI/CD, Copilot, JetBrains AI, etc.).
10. AI implementation features, patterns, and methodology.
11. AI implementation features.
12. Capabilities of custom AI-based solutions for solving unique problems.
13. Joint case study analysis to analyze potential AI applications in non-standard tasks.
14. Metrics for AI application in software development.
15. Joint case study analysis to develop practical skills in selecting and calculating metrics (practical).
16. Review of methods for maintaining data privacy.
17. Review of limitations of AI application.
18. Joint discussion – analysis of results, summing up.

Ask question