Course Outline
Training Program
- Fundamentals of LLM and agent operation in developer work
What is an LLM from a developer's perspective
Differences between a language model, a code assistant, and a coding agent
How models analyze context, code, instructions, and interaction history
LLM limitations
The role of the developer as the decision-maker for technical choices
Typical LLM applications 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 preparing the development environment
Proper repository setup for AI work
Guidelines for preparing project structure, documentation, and agent instructions
Working with the terminal, IDE, and repository in an AI-assisted mode
Controlling the scope of changes and minimizing the risk of unwanted modifications - Agents.md, CLAUDE.md, and skills.md as modules organizing agent work
What agent markdown files are and their main role in LLM work
Difference between a one-time prompt and a persistent project instruction
How to create skills describing coding, testing, and documentation standards
Organizing skills for different types of tasks
Examples of good and bad agent instructions - Plan Mode in practice
What is Plan Mode and when to use it
Planning before implementing code changes
Analyzing risks, dependencies, and potential side effects
Translating the plan into specific repository actions
Iteratively guiding 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 with LLM assistance
Identifying entry points, dependencies, and logic flow
Executing realistic team tasks using an agent
Refactoring a code snippet for readability and maintainability
Generating or supplementing tests
Updating technical documentation and README - Subagents in practice
What subagents are and when to delegate tasks
Dividing work among agents:
Designing the scope of subagent responsibilities
Parallel work and controlling result consistency
Applying subagents in larger refactoring tasks - MCP as a way to extend agent capabilities
What is MCP and its role in AI tool work
MCP client, server, and context sources
Connecting the agent with additional tools, data, and workflows
Examples of MCP applications in developer work - Safety, quality, and responsibility in AI work
Risk of code, data, business logic, and architecture information leaks
Principles for working with production code and critical security segments
Verifying changes generated by AI
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 workflow
Minimal set of practices to implement after the training
Requirements
Programming experience, knowledge of Git basics, and the ability to navigate a code repository are recommended. Participants should know at least one programming language at a level that allows reading and modifying code; examples will be shown using Python. The training does not require prior experience with Claude Code, Codex, Cursor, or MCP, though basic familiarity with terminal usage will be helpful.
Target Group
· Developers at an intermediate to advanced level;
· Developers working with existing, complex, or poorly documented repositories;
· Members of software development teams who wish to organize 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 use LLMs for code analysis, refactoring, documentation, testing, and automating technical tasks.
