Responsible AI in agentic software development

Responsible AI in agentic software developmentRAIASDCYCydrillCY-RAIASD1.0<ul> <li>Understand various aspects of responsible AI</li><li>How to use generative AI responsibly in software development</li><li>Prompt engineering for optimal outcomes</li><li>How to apply generative AI throughout the SDLC</li><li>The challenges in using agentic GenaI</li></ul><p>General development</p><p>All people involved in using agentic AI tools in software development</p><h4>A brief history of Artificial Intelligence</h4><ul> <li>The origins of AI</li><li>Neural networks and “probability engines”</li><li>Early ML coding tools</li><li>The AI coding revolution of the 2020s</li></ul><h4>Responsible AI</h4><ul> <li>What is responsible AI?</li><li>Accountability and transparency</li><li>Mitigation of harmful bias</li><li>Validity and reliability</li><li>Demonstration – Experimenting with validity and reliability in Copilot</li><li>Explainability and interpretability</li><li>Safety, security, privacy and resilience</li><li>Security and responsible AI in software development</li></ul><h4>Using GenAI responsibly in software development</h4><ul> <li>LLM code generation basics</li><li>Basic building blocks and concepts</li><li>Prompt templating</li><li>System prompts in AI-driven coding</li><li>Can AI… boost your productivity?</li><li>Can AI… take care of the ‘boring parts’?</li><li>Can AI… be more thorough?</li><li>Reviewing generated code – the black box blues</li><li>The danger of hallucinations</li><li>The effect of GenAI on programming skills</li><li>Some further long-term effects of using GenAI</li><li>Where AI code generation doesn’t do well</li><li>Prompt engineering <ul> <li>Why is a good prompt so important?</li><li>Establishing the context for generative AI</li><li>Zero-shot, one-shot, and few-shot prompting</li><li>Reasoning-based prompt engineering, chain-of-thought</li><li>Demonstration – Experimenting with prompts in Copilot</li><li>Enforcing and following token limits</li><li>Prompt patterns <ul> <li>Prompt patterns and prompt priming</li><li>The 6 categories of prompt patterns</li></ul></li><li>Some further prompting approaches <ul> <li>Least-to-Most and Self-Planning: decomposition of complex tasks</li><li>Demonstration – Task decomposition with Copilot</li><li>Unit tests, TDD and GenAI</li><li>Demonstration – Test-based code generation with Copilot</li></ul></li></ul></li><li>Integrating generative AI into the SDLC <ul> <li>Using GenAI beyond code generation</li><li>Using AI during requirements specification</li><li>Prompt patterns for requirements capturing</li><li>Prompt patterns for software design</li><li>Demonstration – Requirements capturing and API design with Copilot</li><li>Using AI during implementation</li><li>Prompt patterns for implementation</li><li>Demonstration – Finding hidden assumptions with Copilot</li><li>Using AI during testing and QA</li></ul></li><li>Agentic software development <ul> <li>Intelligent agents and GenAI <ul> <li>How is agentic coding different?</li><li>The Model Context Protocol (MCP)</li><li>Capabilities of MCP agents</li><li>Agentic integration in IDEs</li></ul></li><li>Agentic development workflow <ul> <li>Code-to-spec and spec-to-code with GenAI</li><li>Automated scaffolding</li><li>Demonstration – Agentic scaffolding with Copilot</li><li>Setting up the runtime environment</li><li>Demonstration – Environment setup with Copilot</li><li>Incremental development</li><li>Demonstration – Incremental development with Copilot</li><li>The role of MCP in Dev(Sec)Ops</li><li>Demonstration – Using MCP in DevOps with Copilot</li></ul></li><li>Pitfalls and best practices <ul> <li>“Vibe coding” and its implications</li><li>Engineering concerns with MCP</li><li>Security concerns of agentic development</li><li>MCP’s effect on the attack surface</li><li>MCP-specific attack vectors</li><li>Demonstration – Attacking agentic Copilot</li><li>Case study – Database leakage via Supabase MCP</li><li>Hallucinations and ‘agentic death spirals’</li><li>Token limits and context</li><li>Context degradation with very large token counts</li><li>Prompt engineering vs context engineering</li><li>Context engineering from a developer’s perspective</li><li>Context document examples</li></ul></li></ul></li></ul><h4>Summary and takeaways</h4><ul> <li>Responsible AI principles in software development</li><li>Generative AI – Resources and additional guidance</li></ul><ul> <li>A brief history of Artificial Intelligence</li><li>Responsible AI</li><li>Using GenAI responsibly in software development</li><li>Summary and takeaways</li></ul><p>A must-have primer for those looking to understand and responsibly adopt agentic GenAI in their software development projects. Building on these foundations, and depending on the technology stack, we suggest continuing with one of the Generative AI courses - see Agentic software development with generative AI in C++/Java/C#/Python. For those working on machine learning solutions, the comprehensive 4-day Machine Learning Security course offers a natural next step.</p>- Understand various aspects of responsible AI - How to use generative AI responsibly in software development - Prompt engineering for optimal outcomes - How to apply generative AI throughout the SDLC - The challenges in using agentic GenaIGeneral developmentAll people involved in using agentic AI tools in software developmentA brief history of Artificial Intelligence - The origins of AI - Neural networks and “probability engines” - Early ML coding tools - The AI coding revolution of the 2020s Responsible AI - What is responsible AI? - Accountability and transparency - Mitigation of harmful bias - Validity and reliability - Demonstration – Experimenting with validity and reliability in Copilot - Explainability and interpretability - Safety, security, privacy and resilience - Security and responsible AI in software development Using GenAI responsibly in software development - LLM code generation basics - Basic building blocks and concepts - Prompt templating - System prompts in AI-driven coding - Can AI… boost your productivity? - Can AI… take care of the ‘boring parts’? - Can AI… be more thorough? - Reviewing generated code – the black box blues - The danger of hallucinations - The effect of GenAI on programming skills - Some further long-term effects of using GenAI - Where AI code generation doesn’t do well - Prompt engineering - Why is a good prompt so important? - Establishing the context for generative AI - Zero-shot, one-shot, and few-shot prompting - Reasoning-based prompt engineering, chain-of-thought - Demonstration – Experimenting with prompts in Copilot - Enforcing and following token limits - Prompt patterns - Prompt patterns and prompt priming - The 6 categories of prompt patterns - Some further prompting approaches - Least-to-Most and Self-Planning: decomposition of complex tasks - Demonstration – Task decomposition with Copilot - Unit tests, TDD and GenAI - Demonstration – Test-based code generation with Copilot - Integrating generative AI into the SDLC - Using GenAI beyond code generation - Using AI during requirements specification - Prompt patterns for requirements capturing - Prompt patterns for software design - Demonstration – Requirements capturing and API design with Copilot - Using AI during implementation - Prompt patterns for implementation - Demonstration – Finding hidden assumptions with Copilot - Using AI during testing and QA - Agentic software development - Intelligent agents and GenAI - How is agentic coding different? - The Model Context Protocol (MCP) - Capabilities of MCP agents - Agentic integration in IDEs - Agentic development workflow - Code-to-spec and spec-to-code with GenAI - Automated scaffolding - Demonstration – Agentic scaffolding with Copilot - Setting up the runtime environment - Demonstration – Environment setup with Copilot - Incremental development - Demonstration – Incremental development with Copilot - The role of MCP in Dev(Sec)Ops - Demonstration – Using MCP in DevOps with Copilot - Pitfalls and best practices - “Vibe coding” and its implications - Engineering concerns with MCP - Security concerns of agentic development - MCP’s effect on the attack surface - MCP-specific attack vectors - Demonstration – Attacking agentic Copilot - Case study – Database leakage via Supabase MCP - Hallucinations and ‘agentic death spirals’ - Token limits and context - Context degradation with very large token counts - Prompt engineering vs context engineering - Context engineering from a developer’s perspective - Context document examples Summary and takeaways - Responsible AI principles in software development - Generative AI – Resources and additional guidance- A brief history of Artificial Intelligence - Responsible AI - Using GenAI responsibly in software development - Summary and takeaways A must-have primer for those looking to understand and responsibly adopt agentic GenAI in their software development projects. Building on these foundations, and depending on the technology stack, we suggest continuing with one of the Generative AI courses - see Agentic software development with generative AI in C++/Java/C#/Python. For those working on machine learning solutions, the comprehensive 4-day Machine Learning Security course offers a natural next step.1 jour750.00750.00750.00750.00750.00