Become a Modern Engineer for the AI-native future

Repository instruction files are becoming the control plane for AI-assisted coding: a short AGENTS.md map, deep docs as the system of record, and task-specific specs for execution.

A gateway that discovers, validates, and delegates to a remote specialist over A2A. Both agents run on Microsoft Foundry and deploy to Azure Container Apps from one image.

Build a support agent workflow in C# with trusted vs untrusted content labels, deterministic tool authorization, approval-gated sensitive actions, and redacted audit storage.

Build a durable local-first agent workflow in C# with narrow roles, deterministic policy-bounded execution, live runtime logs, and restart-safe state, using home repair as the concrete example.

Build a grounded local-first multi-agent research workflow in C# with deterministic retrieval, narrow agent roles, structured JSON contracts, and final citation validation.

Build a local-first inference gateway with exact cache keys, per-tenant budget reservation, sync-vs-batch routing, deferred execution, and measurable cost controls in C#.

Build a local-first prompt experiment harness with frozen eval cases, deterministic scoring, live LM Studio execution, and evidence-based prompt promotion.

Build a local-first inference gateway with deterministic rate limiting, queue backpressure, retries, timeouts, circuit breakers, fallback routing, and measurable runtime reliability metrics.

Build a local-first memory architecture with deterministic write guardrails, bounded retrieval, context budgeting, and evidence-only answers.

Build a deterministic context budgeting pipeline that prioritizes required evidence, enforces token limits, and composes inspectable prompts.