Building the Trust Layer for the Agentic Era
summary
Shows a concrete roadmap: Sonar deploys context augmentation that serves code, docs, tickets and live SonarQube state to agents, an MCP server to connect IDEs and cloud cursors, and a new SonarQube architecture engine that detects architecture drift as issues. Benchmarks claim ~66% fewer issues with contextualized agents and a remediation agent that achieves 79.2% automatic fix success in beta. Demonstrates agentic analysis hooks producing cleaner PRs and an automated guide→verify→solve loop.
AI-written summary · about
chapters (30)
- 0:00Introduction
- 0:17Adapting Faster in the Era of AI-Powered Development
- 1:01Copilot Coding vs Agent-Driven Development Models
- 1:23Why AI Coding Agents Need Planning and Context
- 2:07The AI Code Generation and Verification Loop
- 2:58The Guide Stage: Providing Context to AI Coding Agents
- 3:37Context Augmentation for AI Software Development Agents
- 4:05What “Context” Means in AI-Assisted Coding
- 4:42Delivering Code Context to AI Developer Tools
- 5:09Static and Dynamic Context for AI Code Generation
- 5:57Benchmark Results: Fewer Bugs and Vulnerabilities with Context
- 6:43MCP Server: Integrating AI Coding Agents with Developer Tools
- 7:30Architecture Drift in AI-Generated Code
- 8:14Introducing SonarQube Architecture for Governance
- 9:28Architecture Discovery and Codebase Dashboards
- 10:01Defining Intended Architecture and Enforcing Rules
- 10:28Detecting Architecture Deviations and Code Quality Issues
- 10:53Guide Stage Recap and Product Roadmap Updates
- 11:26Verify Stage: Checking AI-Generated Code with Analysis
- 12:12Agentic Code Analysis for Faster Verification
- 12:49Running Sonar Verification in the Developer Workflow
- 13:30Sending Code Quality Feedback to AI Coding Agents
- 14:43Creating Clean Pull Requests with Automated Verification
- 15:11Solve Stage: AI Remediation Agents Fixing Code Issues
- 15:54Remediation Agent Benchmark Results
- 16:50Enabling and Assigning a Code Remediation Agent
- 17:48Monitoring AI Agent Fixes and Pull Request Changes
- 18:27Automatic Code Remediation After Failed Quality Gates
- 19:08Guide–Verify–Solve Workflow for AI Development
- 19:28Final Thoughts: Responsible AI-Assisted Software Development
related talks
29mBuilding Trust in AI Agents
28mAgentic Development at Enterprise Scale: RAG, MCP & Guardrails
44mArchitects of Trust: Redefining Quality Leadership in an Agentic World
26mArchitecting Agentic Development Platforms
30mArchitecting Agentic Development Platforms
27mThinking in platforms: Thrive in the agentic era
44mAI Architecture Thinking: From Idea to Impact
35mThe Agentic AI Engineer