Course Outline
Training Program
- Fundamentals of LLMs and agents in developer work
What is an LLM from a programmer's perspective
Differences between a language model, code assistant, and coding agent
How the model analyzes context, code, instructions, and interaction history
Limitations of LLMs
The developer's role as the decision-maker for technical matters
Typical applications of LLMs in development - Setting up Claude Code, Codex, and Cursor for daily AI work
Overview of tools used in agent-based code work
Claude Code, Codex, and Cursor – differences in workflow and typical use cases
Installation, configuration, and preparation of the development environment Correct repository setup for AI work
Principles for preparing project structure, documentation, and instructions for the agent
Working with the terminal, IDE, and repository in an AI-supported mode Controlling the scope of changes and minimizing the risk of unintended modifications - Agents.md, CLAUDE.md, and skills.md as modules organizing agent work
What agent markdown files are and the main role they play in LLM work
Difference between a one-time prompt and persistent project instructions
How to create skills describing coding, testing, and documentation standards Organizing skills for different types of tasks
Examples of good and bad instructions for agents - Plan Mode in practice
What Plan Mode is and when to use it
Planning before executing code changes Analyzing risks, dependencies, and potential side effects Translating a plan into concrete actions in the repository Iterative guidance of the agent
Working with larger changes
Evaluating the quality of the plan generated by AI - Practical exercises on open repositories
Onboarding to an unknown codebase using an LLM Identifying entry points, dependencies, and logic flow Executing realistic team tasks using an agent Refactoring a code fragment for readability and maintainability Generating or supplementing tests Updating technical documentation and README - Sub-agents in practice
What sub-agents are and when to delegate tasks Division of work between agents:
Designing the scope of responsibility for a sub-agent Parallel work and control of result consistency Applying sub-agents in larger refactoring tasks - MCP as a way to extend agent capabilities
What MCP is and its role in AI tooling MCP client, server, and context sources Connecting the agent with additional tools, data, and workflows Examples of MCP applications in a programmer's work - Security, quality, and responsibility in AI-assisted work Risks of code, data, business logic, and architecture information leaks Rules for working with production code and critical security fragments Verification of AI-generated changes AI in the code review process Best practices for implementing agent-based code work in an organization
- Summary and practical implementation Key principles for effective developer work with LLMs How to transfer training knowledge into daily workflows Minimum set of practices to implement after the training
Requirements
Programming experience is recommended, along with knowledge of basic Git and the ability to navigate a code repository. Participants should know at least one programming language at a level that allows reading and modifying code; examples will be demonstrated using the Python language. Previous experience with Claude Code, Codex, Cursor, or MCP is not required, though basic familiarity with working in a terminal will be helpful.
Target Audience
· Developers at intermediate and advanced levels;
· Developers working with existing, complex, or poorly documented repositories;
· Software development team members who wish to organize and structure AI usage practices in their daily work;
· Tech leads and senior developers responsible for code quality, code review, and selecting developer tools;
· Individuals wishing to consciously utilize LLMs for code analysis, refactoring, documentation, testing, and automation of technical tasks.
