Stack & tools
Languages and frameworks are still in the toolkit, but they sit in the background. What matters now is knowing what good software looks like, and coaching yourself, your agents, and your delivery pipeline toward steady improvement in an agentic workflow.
Architecture patterns
- Enterprise architecture, platform direction, and NFRs in regulated domains
- Agentic pipelines with schema-bound outputs, verification loops, and auditability
- Shift-left testing and regression control for non-deterministic AI components
- Angular schematics and engineering standards at team scale
Languages & frameworks
- C# / .NET (services, installers, background sync)
- TypeScript / Angular (product UIs)
- Swift, Objective-C, Kotlin, Java for mobile and JVM application stacks
- Astro, PHP, SQL (selected by project context)
Cloud & infrastructure
- Microsoft Azure (AI Foundry, Functions, Cosmos DB, Blob Storage)
- Static hosting and CI deploy patterns
Development tools
- Git for branching, review, and release workflows
- Visual Studio, Visual Studio Code, Xcode, Android Studio: IDEs by stack and platform
- Cursor and Claude for agent-assisted development alongside conventional tooling
AI/ML tools
- Azure AI Foundry and model-agnostic agent frameworks
- LLM orchestration with deterministic verification companions
- Cursor-style rules, skills, and agent workflows for engineering quality
Learning next
- Deeper agent evaluation and continuous improvement loops for production AI
Skills are table stakes. The proof is in the case studies and experience.