Course Outline
Session 1: Reproducible Methodology and Tool Map (2h)
-
Module 1: From personal flow to reproducible methodology
-
Explicit team contracts: Sustaining the SDD flow at a coordinated collective scale.
-
Specification lifecycle: Refinement, validation, and traceability under high pressure.
-
Reproducibility guarantee with AI: Structural consistency (same input → same output) in multi-agent environments.
-
-
Module 2: Multi-tool operational map
-
Local context control: Configuration of environments and guide files (
CLAUDE.mdvs.copilot-instructions.mdvs.OpenCode). -
Instructional guidance structure: Equivalencies and discrepancies between Rules, AI Contexts, system instructions, and memory.
-
Multi-agent synchronization: Maintaining a single source of functional truth with heterogeneous tools.
-
Session 2: Precision in Artifacts and Calibration (2h)
-
Module 3: Precision in artifacts: The differences that matter
-
Specification boundaries: Technical differences and agent behavior regarding Functional Specs vs. Technical Specs.
-
Task granularity and sizing: Impact of massive tasks (code breaking) vs. tiny tasks (loss of context).
-
Comparative response analysis: Course adjustments in Copilot, Claude, and OpenCode for the same specification.
-
Production antipatterns: Live demonstration of recurrent failures and their impact on generated code.
-
-
Guided Practice: Work on real specifications (30 min)
-
Active analysis and real-time diagnosis of specifications provided by the team to detect ambiguities.
-
Session 3: Full Lifecycle and Change Management (2h)
-
Module 4: Refinement, validation, and mid-flight changes
-
Iterative refinement techniques: Transitioning from vague needs to validated specifications without writing base code.
-
Operational feasibility checklist before delegating tasks to an agent.
-
Mid-course modifications: Managing technical scope variations midway through the cycle without corrupting previous tasks.
-
Control of structural deviations: Procedures to regain control when the agent diverges from the objective.
-
-
Module 5: Multi-agent patterns and MCPs in a corporate environment
-
Limits of the agentized paradigm: Formal criteria to avoid excessive structural complexity without value return.
-
Advanced development patterns: Hierarchical orchestration, parallel agents, and human control gates.
-
MCP Ecosystem (Model Context Protocol): Corporate implementation with Filesystem, Git, and Jira (Direct demonstration).
-
Session 4: Quality, Security, and Real Case Resolution (2h)
-
Module 6: Agent-driven TDD and quality pipeline
-
Specification as a test contract: Transformation of acceptance criteria into automated tests by AI.
-
Automation in PR and CI/CD: What to automate and what decisions to retain under strict human review.
-
Continuous integration differences: Coupling methodologies for GitHub Copilot, Claude, and OpenCode.
-
-
Module 7: Security, intellectual property, and real limits
-
Data security policies: Protection of sensitive information and advantages of corporate plans (Enterprise/Team).
-
Intellectual property: Regulatory scenario, authorship, and responsibilities in AI-assisted development.
-
Organizational guardrails: Ethical and internal frameworks to enhance delivery without compromising technological assets.
-
-
Practical Resolution: Real use cases from the team (30 min)
-
Solution in the team’s code: Direct work on the repository and issues provided by the participants.
-
Requirements
-
Daily and established practice in Software-Driven Development (SDD - Software-Driven Development).
-
Prior experience and active use in their workflow of at least one of the following tools: GitHub Copilot, Claude, or OpenCode.
-
Solid knowledge in software architecture, specification management (functional and technical), and continuous integration workflows (CI/CD).
-
Essential requirement: Willingness to provide real use cases and code examples from the team before the start of the training.
Target Audience:
- A consolidated team of 20 professionals with daily SDD practice (Senior Developers, Tech Leads, and Software Architects who already use AI in their day-to-day work and seek to optimize, standardize, and scale their workflow).
-
Essential requirement: Willingness to provide the trainer with real use cases and code examples from the team before the start of the training for a preliminary expectation management meeting.
Custom Corporate Training
Training solutions designed exclusively for businesses.
- Customized Content: We adapt the syllabus and practical exercises to the real goals and needs of your project.
- Flexible Schedule: Dates and times adapted to your team's agenda.
- Format: Online (live), In-company (at your offices), or Hybrid.
Price per private group, online live training, starting from 2900 € + VAT*
Contact us for an exact quote and to hear our latest promotions
Testimonials (4)
What appears most useful is the fact that both theory and practical demonstrations are shared, which helps ground the knowledge and reveals its real-world utility.
Noelia Iglesias Martinez - Aplicaciones y Tratamientos de Sistemas S.A (knowmad mood)
Course - IA Generativa y SDD Avanzado
Machine Translated
the MCPs
Gustavo Lopez Garcia - Aplicaciones y Tratamientos de Sistemas S.A (knowmad mood)
Course - IA Generativa y SDD Avanzado
Machine Translated
the automation flow of the agents
Antonio Perez Caballero - Aplicaciones y Tratamientos de Sistemas S.A (knowmad mood)
Course - IA Generativa y SDD Avanzado
Machine Translated
AGENTS.MD concepts
Eduardo Palomares Medina - Aplicaciones y Tratamientos de Sistemas S.A (knowmad mood)
Course - IA Generativa y SDD Avanzado
Machine Translated
