Practical AI Adoption for Technical Teams

The problem this course solves

Most technical teams already use AI occasionally: a one-off prompt in ChatGPT, an autocomplete suggestion. Very few have adopted it as a real part of their daily workflow. That happens because:

• There is no proven working framework, so everyone reinvents their own process by trial and error.
• The learning curve for tools like coding agents is walked alone, without a map of which patterns work and which are common mistakes.
• Adoption stays at "use the tool when you remember it" instead of becoming a structural part of the work cycle.
• There is no organized way to structure AI-assisted projects beyond the individual prompt.

Jorge Domínguez has spent more than two years working daily with Claude Code, Claude AI and the AI-assisted development ecosystem on real production projects. This course distills that experience into a concrete working framework participants can start using the next day — not theory or repackaged official documentation.

Who it is for

Designed for technical teams who already write, review or maintain software and want to make AI a structural part of their work rather than an occasional experiment.

• Software developers with hands-on experience in any programming language.
• Software engineers, development analysts and solution architects.
• DevOps, QA and related professionals interested in AI-assisted automation.
• Whole technical teams looking to adopt AI consistently rather than in scattered ways.

Prerequisites

• Basic programming knowledge in any language.
• Familiarity with a development environment (IDE, terminal or code editor).
• No prior experience with AI tools is required, but a willingness to experiment is.

What participants take away

• A ready-to-use operational baseline, not an introduction: their own system of technical prompts.
• An AI agent workflow applied to real development tasks: code generation, debugging, refactoring and documentation.
• Applied understanding of the BMAD framework for structuring AI-assisted projects.
• A practical introduction to the MCP protocol for connecting agents to real tools and environments (IDEs, repositories, cloud services).
• End-to-end experience solving a case of their own with everything above integrated.

Learning objectives

By the end of the course, participants will be able to:

1. Understand the developer's new role in an AI-assisted development context.
2. Build effective technical prompts for code generation, debugging, refactoring and documentation.
3. Use development agents (such as Claude Code) and conversational assistants (such as Claude AI) as copilots inside real engineering workflows.
4. Apply the principles of the BMAD framework to structure AI projects in an organized way.
5. Understand the MCP protocol and its integration possibilities with development tools and environments.
6. Integrate AI tools into the daily workflow sustainably and progressively.
7. Solve an end-to-end practical case applying everything covered in the course.

Course content

1. Introduction to AI adoption for technical teams

   An overview of AI applied to software development and how the developer's role changes. Distinguishes occasional AI use from adopting it as part of the workflow, and introduces a product mindset applied to AI development.

2. Practical prompting for engineering and development

   Fundamentals of technical prompting: how to structure context, instructions, constraints and output criteria. Includes advanced techniques — multi-step prompts, chain-of-thought, few-shot — applied to code generation, debugging, refactoring and documentation. Closes with a personal library of technical prompts.

3. AI-assisted development with coding agents

   Practical use of development agents and conversational assistants as copilots inside a realistic engineering workflow: building and improving features, refactoring existing code, generating technical documentation and supporting debugging.

4. The BMAD framework applied

   Introduces BMAD as a structure for organizing AI-assisted development work: its phases and the logic of discovery, design, execution and continuous improvement. Participants turn a real technical need into an organized BMAD work sequence.

5. Connecting agents to tools: introduction to MCP

   Introduces the value of connecting agents to real development tools and environments, and presents MCP as the key protocol for extending agent capabilities. Reviews concrete integration cases: IDEs, repositories, cloud services and project management.

6. Integrating hands-on workshop

   Consolidates everything into a guided case: technical prompting, assisted development, structuring with BMAD and a view of tool integration. Closes with sharing results and adoption recommendations.

Methodology

Duration and format

• Duration: 24 hours.
• Suggested format: 6 sessions of 4 hours.
• Delivery: virtual, by videoconference.
• Tools: Claude AI and Claude Code as the core (an active Claude Pro subscription or equivalent is required); Node.js installed for coding-agent setup.

About the instructor

Jorge Domínguez is a systems engineer with more than 18 years of experience in software development, solution architecture and technology innovation. He holds an MBA from Universität Leipzig, Germany, focused on innovation and technology adoption, and is a graduate of the MIT xPRO executive program in AI product and service design and development (2025). He is certified in Generative AI Fundamentals by Databricks and in Teaching the AI Fluency Framework by Anthropic.

He works actively as an AI Solution Architect on AI-first enterprise integration projects, implementing workflows with MCP, AI agents and cloud-native architectures on AWS. Specialized in Claude Code, Claude AI, Cursor and ChatGPT for agent-assisted development, he uses these tools daily on real production projects. He has trained more than 60 companies and institutions — including Bancolombia, Comfandi and Universidad de los Andes — in AI adoption at both technical and organizational levels.

Next step

If your organization has not yet defined its mandate and usage rules, start with AI Governance — 4Ds Framework. To take this technical capability into a governed proof of concept, see the AI Adoption & Governance Sprint. You can also read our practical Claude Code guide.