Week 04 — Orchestrator-workers, AI-assisted dev, stack — interview-1¶

At a glance

Phase 1: Foundations · Starts: Mon 19 Oct 2026 · Planned: 16h 45m

Build: Capstone: orchestrator-workers 'draft postmortem' with evaluator-optimizer loop + 10 golden examples

Interview checkpoint: interview-1

Run the full mock loop and score it with the rubric.

By day¶

System Design · 45 min · Interview framework refresh: practise requirements + estimation opening for 'design Pastebin' in 10 min, recorded → Interview framework & back-of-envelope estimation · resource w04-sd-1
DSA · 25 min · Valid Parentheses + Min Stack → Stack & monotonic stack · resource w04-dsa-1
Python · 20 min · Rep: asyncio.Semaphore + bounded queue for backpressure on concurrent LLM calls → asyncio in depth: TaskGroups, cancellation, backpressure w04-py-1
Communication · 30 min · Grammar: Conditionals and modals for hedging and softening → Week 04 drills · day 1 w04-comm-1

Agentic AI · 60 min · AI-assisted development: write AGENTS.md for the capstone repo (commands, conventions, test policy) and add a Claude Code skill for 'add a tool' → AI-assisted development: coding agents, AGENTS.md, skills · resource w04-ai-1
DSA · 30 min · Evaluate Reverse Polish Notation → Stack & monotonic stack · resource w04-dsa-2
Communication · 30 min · Vocabulary: Diplomatic and hedging language → Week 04 drills · day 2 w04-comm-2

DSA · 25 min · Daily Temperatures (monotonic decreasing stack) → Stack & monotonic stack · resource w04-dsa-3
Architecture · 45 min · SOLID & refactoring at scale: refactor the capstone tool loop (SRP, DIP via ports), note code smells found → SOLID, refactoring & code quality at scale · resource w04-arch-1
Python · 20 min · Rep: async context managers (contextlib.asynccontextmanager) for a pooled client; generator-based pipelines → Iterators, generators & context managers w04-py-2
Communication · 30 min · Speaking drill: Cutting filler words (um, like, you know) → Week 04 drills · day 3 w04-comm-3

Agentic AI · 60 min · Context engineering lab: just-in-time retrieval (tool fetches runbook) vs preloaded context — compare token usage and answer quality on 10 cases → Context engineering · resource w04-ai-2
Java/Spring AI · 30 min · Java 25 features: compact source files, flexible constructor bodies, module imports, scoped values (final) — 4 micro-examples → Modern Java 21→25: records, sealed types, patterns · resource w04-java-1
Communication · 30 min · Idioms & phrasal verbs: Negotiation: meet halfway, off the table → Week 04 drills · day 4 w04-comm-4

System Design · 45 min · Scalability patterns recap: stateless services, horizontal scaling, queues for load levelling — map to the capstone → Scalability fundamentals & latency numbers · resource w04-sd-2
DSA · 25 min · Car Fleet (sort + stack) → Stack & monotonic stack · resource w04-dsa-4
Python · 20 min · Rep: FastAPI StreamingResponse / SSE for token streaming with client-disconnect handling → FastAPI for production w04-py-3
Communication · 30 min · Writing: Editing for concision: cut 30 percent → Week 04 drills · day 5 w04-comm-5

Agentic AI · 120 min · Capstone build: orchestrator-workers for 'draft postmortem' (timeline, impact, root cause workers) + evaluator-optimizer critique loop; commit 10 golden examples as seed eval set → Agent & workflow patterns (Building Effective Agents) · resource w04-ai-3
DSA · 30 min · Timed coding mock (interview-1): Largest Rectangle in Histogram in 30 min, narrate trade-offs → Stack & monotonic stack · resource w04-dsa-5
Java/Spring AI · 30 min · Port the /triage endpoint to Spring AI (ChatClient + record output) and compare with Python implementation → Spring AI 2.0 fundamentals: ChatClient, advisors, structured output w04-java-2
Communication · 30 min · Record & shadow: Cutting filler words (um, like, you know) → Week 04 drills · day 6 w04-comm-6

System Design · 45 min · Timed 45-min SD mock (interview-1): Pastebin / URL shortener variant with custom aliases + expiry; self-score → URL shortener w04-sd-3
Architecture · 30 min · Write ADR 004 for capstone: structured-output validation strategy (schema-constrained decoding + bounded retry) → The architect role & trade-off thinking · resource w04-arch-2
Staff+ · 30 min · List 3 candidate 'Staff projects' at work or in the capstone with scope, stakeholders, and why they are Staff-level → Staff archetypes & operating at Staff+ · resource w04-staff-1
Communication · 30 min · Soft skills + weekly review: Disagreeing respectfully — checkpoint comm-1: re-record and compare against the baseline rubric → Week 04 drills · day 7 w04-comm-7
Review · 60 min · interview-1: full mock — SD + coding + AI SD + behavioral, score with the unified interview rubric and log gaps into next week's plan → Interview framework & back-of-envelope estimation w04-rev-1