AI-Assisted Coding in Software Development

TRAINING DESCRIPTION

This training takes a pragmatic, buzzword-free look at agentic software development: what coding agents can and cannot do today, and how to put them to work across the entire software lifecycle. Instead of chasing demos, participants learn to design reliable, repeatable AI-assisted workflows—from requirements gathering and design through implementation, deployment, and release.

The course covers today’s leading coding agents (Claude Code, Codex, OpenCode) and their core building blocks—skills, subagents, hooks, verification loops, and scheduled or background work. Participants map their own team’s software development process into AI-assisted workflows, then go a step further into AI-native development, where background agents and automation carry a growing share of the work. By the end, participants can introduce coding agents into a real codebase and team with confidence, quality, and control.

This training is for software developers, tech leads, and architects who want to move beyond ad-hoc prompting and build a dependable, team-wide practice of AI-assisted development.

Order dedicated training

BASIC PROGRAM

  • Module 1: Capabilities and Limitations of Agentic Software Development
  • Module 2: Coding Agents
  • Module 3: Building Workflows
  • Module 4: AI-Assisted Requirements Gathering, Analysis, and Documentation
  • Module 5: AI-Assisted Software Design and Architecture
  • Module 6: AI-Assisted Implementation
  • Module 7: AI-Assisted Deploy and Release
  • Module 8: Advanced Topics
  • Module 9: A Step Further — AI-Native Software Development

DETAILED PROGRAM

Module 1: Capabilities and Limitations of Agentic Software Development

  • What agentic software development is—and what it is not
  • Current state of the art in companies: where agents already deliver value
  • Honest limitations: reliability, context, cost, and trust boundaries
  • Future options and where the field is heading

Module 2: Coding Agents

  • Overview of current coding agents (Claude Code, Codex, OpenCode)
  • Main features of coding agents and how they differ
  • Skills, subagents, and hooks as reusable building blocks
  • Verification loops and reliability engineering for trustworthy output
  • Scheduled work and background agents

Module 3: Building Workflows

  • Workflow anatomy: the elements that make a workflow repeatable
  • Steps and actions, services, knowledge, and gates
  • From a real-life workflow to an AI-assisted workflow design
  • Mapping the software development workflow for your team

Module 4: AI-Assisted Requirements Gathering, Analysis, and Documentation

  • Turning conversations and raw notes into structured requirements
  • Analyzing scope, edge cases, and ambiguities with agents
  • Generating and maintaining living documentation
  • Keeping humans in the loop for validation and sign-off

Module 5: AI-Assisted Software Design and Architecture

  • Exploring design options and trade-offs with agents
  • Producing architecture drafts, diagrams, and decision records
  • Reviewing designs for consistency, risks, and constraints
  • Keeping design artifacts synchronized with the codebase

Module 6: AI-Assisted Implementation

  • Driving implementation through agents while staying in control
  • Test-first and verification-driven coding with agents
  • Reviewing, refactoring, and hardening generated code
  • Managing context, scope, and surgical changes

Module 7: AI-Assisted Deploy and Release

  • Automating build, deploy, and release steps with agents
  • Release checks, gates, and rollbacks
  • Observability and post-release verification
  • Balancing automation with safe human control points

Module 8: Advanced Topics

  • Multirepo and monorepo support
  • Configurable workflows for different projects and contexts
  • Working in a team: shared conventions, skills, and guardrails

Module 9: A Step Further — AI-Native Software Development

  • Background agents that carry ongoing work
  • Automation across the development lifecycle
  • Designing a team practice where AI-native development is the default

KEY TAKEAWAYS

  • Understand what agentic software development can and cannot do today—without the hype.
  • Work fluently with modern coding agents and their skills, subagents, hooks, and verification loops.
  • Design AI-assisted workflows and map your team’s development process onto them.
  • Apply agents across the full lifecycle: requirements, design, implementation, deploy, and release.
  • Take the step toward AI-native development with background agents and automation.