Presentations
Technical deep-dives for practitioners on specific capabilities
Under the hood diagnostics. Chat Debug View, customization diagnostics, and extension logs for systematic troubleshooting.
AI assistance at the terminal. Natural language to shell commands, git operations, and scripting workflows.
Rapid Azure prototyping and deployment. Infrastructure becomes conversational — AI that knows what's running in your subscriptions.
Accelerating PR velocity with AI-powered review. 40-60% faster cycle times, custom compliance rulesets, and measurable ROI.
AI assistance beyond the IDE. PR reviews in browsers, issue triage on mobile, with full repository customizations.
Latest Copilot features in VS Code v1.108-v1.110. Agent Plugins, /create-* commands, Explore subagent, agentic browser tools, and /yolo mode.
The 4 configuration primitives. Instructions, prompts, skills, and agents — make Copilot understand your codebase better.
Programmable governance for AI agents. Real-time prevention, audit trails, and compliance enforcement at execution time.
Cross-session context persistence. Store preferences, project context, and workflow patterns that persist across all conversations.
Embedding AI agents in your applications. Build custom tools, bots, and automation with programmatic Copilot access.
Rich UI in chat responses. Transform chat from text-only to interactive visual experiences with component-based MCP tools.
From subagents to autonomous teams. The complete stack: context isolation, role specialists (Squad), background agents + worktrees, and multi-model deliberation (AgentCouncil).
Merged into Building Agent Systems. Subagent delegation, context isolation, and phased workflows are now covered there.
Merged into Building Agent Systems. Background agents + Git worktrees for autonomous parallel execution are now covered there.
The Agent Client Protocol for universal AI integration. Use Copilot's agentic capabilities in any editor, tool, or workflow — not just VS Code.
Repository automation with AI agents. Write workflows in Markdown, compiled to secure GitHub Actions with safe-output validation.
From "assign to copilot" to full SDLC transformation. The 5-phase incremental path with 4-workflow architecture and historical context search.
Level 5 infrastructure for AI velocity. Rewire repositories, PR workflows, and CI/CD to scale from 2-3 features/week to 10-15/day with autonomous agents.
Scaling Copilot across organizations. From team silos to standardized foundations with measurable ROI.
Strategic thought leadership for technical decision-makers
No instruments, no delivery. Why enterprises can't fly on AI autopilot without governance, compliance, and audit trails.
The $2/hour engineer. Making the business case for AI agents—what work moves to $2/hour labor?
The invisible 80%. Beyond code generation—how agents transform issue triage, compliance, documentation, and knowledge transfer.
Hands-on training modules with personas, exercises, and metrics
The Challenge: Build FanHub in 8 hours. Meet the team, understand the messy codebase, embrace the constraint.
The magic file. Stop Copilot from guessing—document your architecture and patterns so AI suggestions are consistent.
From implementation to orchestration. Let AI research dependencies and plan multi-file features before coding.
Build reusable prompts then graduate them to agent skills. Turn repeated patterns into team-wide functions and packaged domain capabilities.
Beyond static files. Connect Copilot to live databases, APIs, and external tools as first-class context sources.
End context-switching overhead. Bundle tools, skills, and instructions into role-based presets you switch instantly.
Companion repositories with working examples and reference implementations
3M LoC C# codebase for stress-testing Copilot on large repos. Breadcrumb navigation, harness benchmarks, and customization patterns at scale.
Intentionally broken workshop starter in 4 languages (Node, .NET, Java, Go). 180+ seeded bugs to fix with Copilot — the messiness is the point.
Web UI for orchestrating Copilot CLI agents across multiple repos via Agent Client Protocol. Broadcast prompts, coalesce results, coordinate cross-repo work.
Multi-agent deliberation framework. Agents debate, vote, and reach consensus on complex decisions through structured council sessions.
Coordinated multi-agent workflows using the Agent Client Protocol. Define agent teams, assign roles, and orchestrate collaborative task execution.