Week 15 — Managed platforms & production checklist¶

At a glance

Phase 4: Production & AI Architecture · Starts: Mon 04 Jan 2027 · Planned: 16h 45m

Build: Capstone: Microsoft Foundry Agent Service spike + production-readiness checklist pass

By day¶

System Design · 45 min · Zero trust + agent identity: workload identity, per-tool scopes, tenant isolation patterns (silo/pool/bridge) → Security: authN/Z, OAuth2/OIDC, zero trust, multi-tenancy w15-sd-1
DSA · 25 min · Swim in Rising Water (Dijkstra / binary search + BFS) → Advanced graphs: Dijkstra, MST, Bellman-Ford · resource w15-dsa-1
Python · 20 min · Rep: py-spy / cProfile flame graph of the ingestion pipeline → Performance & profiling w15-py-1
Communication · 30 min · Grammar: Comparatives, quantifiers and precise comparisons → Week 15 drills · day 1 w15-comm-1

Agentic AI · 60 min · Managed agent platforms: Microsoft Foundry Agent Service (Responses API) vs Bedrock AgentCore — map capstone components to each → Managed platforms: Microsoft Foundry, Bedrock AgentCore · resource w15-ai-1
DSA · 30 min · Reconstruct Itinerary (Hierholzer) → Advanced graphs: Dijkstra, MST, Bellman-Ford · resource w15-dsa-2
Communication · 30 min · Vocabulary: Data storytelling vocabulary → Week 15 drills · day 2 w15-comm-2

DSA · 25 min · Climbing Stairs + Min Cost Climbing Stairs → Dynamic programming 1-D · resource w15-dsa-3
Architecture · 45 min · Evolutionary architecture: fitness functions — turn eval thresholds, latency budgets and dependency rules into CI checks → Evolutionary architecture & fitness functions w15-arch-1
Python · 20 min · Rep: memray/scalene memory profile of embedding batch job → Performance & profiling w15-py-2
Communication · 30 min · Speaking drill: Narrating a demo → Week 15 drills · day 3 w15-comm-3

Agentic AI · 60 min · Production checklist: evals gate, observability, guardrails, cost caps, fallbacks, rollback, data retention, incident runbook — audit the capstone → Shipping LLM features to production: the checklist · resource w15-ai-2
Java/Spring AI · 30 min · Secure the Spring AI MCP server with OAuth2 resource server (Spring Security) per MCP authorization spec → Tool calling & MCP servers in Spring AI · resource w15-java-1
Communication · 30 min · Idioms & phrasal verbs: Data idioms: tip of the iceberg, moving target → Week 15 drills · day 4 w15-comm-4

System Design · 45 min · Quotas for LLM platforms: token-based rate limits per tenant, fairness, burst handling → Rate limiting & quotas w15-sd-2
DSA · 25 min · House Robber + House Robber II → Dynamic programming 1-D · resource w15-dsa-4
Python · 20 min · Rep: refactor a service layer to use the message bus for side effects → Architecture patterns in Python (repository, UoW, message bus) · resource w15-py-3
Communication · 30 min · Writing: Tech-debt proposal → Week 15 drills · day 5 w15-comm-5

Agentic AI · 120 min · Capstone build: run the triage agent on Foundry Agent Service (or MS Agent Framework) against the same eval set; compare quality, latency, cost, ops burden → Managed platforms: Microsoft Foundry, Bedrock AgentCore · resource w15-ai-3
DSA · 30 min · Longest Palindromic Substring → Dynamic programming 1-D · resource w15-dsa-5
Java/Spring AI · 30 min · Spring Boot 4 production settings: graceful shutdown, virtual threads flag, actuator probes for Container Apps → Spring Boot 4 & Spring Framework 7 w15-java-2
Communication · 30 min · Record & shadow: Narrating a demo → Week 15 drills · day 6 w15-comm-6

System Design · 45 min · Written design: LLM evaluation & observability platform — trace ingestion, datasets, judges, online evals, dashboards → Design an LLM evaluation & observability platform · resource w15-sd-3
Architecture · 30 min · Write ADR 015 for capstone: architectural fitness functions in CI (eval score, p95, import-linter rules) → Evolutionary architecture & fitness functions · resource w15-arch-2
Staff+ · 30 min · Artifact: tech-debt proposal — inventory, cost of delay, prioritised paydown plan, how you'd get it funded → Managing technical debt & quality w15-staff-1
Communication · 30 min · Soft skills + weekly review: Difficult conversations → Week 15 drills · day 7 w15-comm-7
Review · 60 min · Explain-it-back: managed vs self-built agent platform trade-offs; flashcards on production checklist items → Shipping LLM features to production: the checklist w15-rev-1