Week 25 — Buffer I — catch-up + specialization choice¶

At a glance

Phase 7: Buffer · Starts: Mon 15 Mar 2027 · Planned: 11h 00m · Light week

Build: Catch up on any unfinished capstone increments; choose a specialization (serving, evals platform, or multi-agent) and scope a 2-week spike

By day¶

System Design · 30 min · Redo weakest SD topic from interview rubric scores → Reliability: retries, backoff, circuit breakers, bulkheads w25-sd-1
DSA · 15 min · Revision (weak pattern): Largest Rectangle in Histogram → Stack & monotonic stack · resource w25-dsa-1
Python · 15 min · Rep: packaging — publish the eval harness as an internal package → Packaging, project layout & monorepos w25-py-1
Communication · 20 min · Grammar: Free choice: your weakest grammar area → Week 25 drills · day 1 w25-comm-1

Agentic AI · 40 min · Catch-up: add Pydantic AI output validators, ModelRetry and usage limits to every sub-agent; re-run evals → Pydantic AI · resource w25-ai-1
DSA · 20 min · Revision: Median of Two Sorted Arrays → Binary search (incl. on answer) · resource w25-dsa-2
Communication · 20 min · Vocabulary: Free choice: your weakest vocabulary area → Week 25 drills · day 2 w25-comm-2

DSA · 15 min · Revision: Merge K Sorted Lists → Linked list · resource w25-dsa-3
Architecture · 30 min · Catch-up architecture reading (Hard Parts / Team Topologies chapter of choice) → Coupling, cohesion & modularity (balanced coupling) w25-arch-1
Python · 15 min · Rep: data tooling — eval analytics notebook with Polars → Data tooling: Polars, DuckDB, Arrow w25-py-2
Communication · 20 min · Speaking drill: Free choice: your weakest speaking area → Week 25 drills · day 3 w25-comm-3

Agentic AI · 40 min · Specialization spike choice: write a 1-page proposal (serving vs evals platform vs protocols) with success criteria → A2A & AG-UI protocols · resource w25-ai-2
Java/Spring AI · 20 min · Catch-up Java task from progress.yml → Agentic patterns with Spring AI w25-java-1
Communication · 20 min · Idioms & phrasal verbs: Idiom review round → Week 25 drills · day 4 w25-comm-4

System Design · 30 min · Read one paper from the list (Dynamo/Spanner/Kafka) and summarise trade-offs → Databases: SQL vs NoSQL, storage engines · resource w25-sd-2
DSA · 15 min · Revision: N-Queens → Backtracking · resource w25-dsa-4
Python · 10 min · Rep: observability — trace sampling config → Observability in Python: structlog, OpenTelemetry w25-py-3
Communication · 20 min · Writing: Free choice: your weakest writing area → Week 25 drills · day 5 w25-comm-5

Agentic AI · 75 min · Specialization build block 1: benchmark pgvector vs Qdrant on the capstone corpus (filtered search, recall@10, p95, ops cost) and record results → Vector databases: pgvector, Qdrant & friends · resource w25-ai-3
DSA · 20 min · Revision: Alien Dictionary (topological sort) → Advanced graphs: Dijkstra, MST, Bellman-Ford · resource w25-dsa-5
Java/Spring AI · 20 min · JVM performance: one improvement measured end to end → JVM performance, GC & Leyden/AOT w25-java-2
Communication · 20 min · Record & shadow: Free choice: your weakest speaking area → Week 25 drills · day 6 w25-comm-6

System Design · 30 min · Written design: redo the lowest-scored mock design → Design an intelligent document processing pipeline w25-sd-3
Architecture · 20 min · Write ADR 025 for capstone: specialization spike architecture → Architecture Decision Records · resource w25-arch-2
Staff+ · 20 min · Tech-debt follow-up: score the capstone's debt after v1.0 and add the top-3 items to the v2 roadmap with cost of delay → Managing technical debt & quality w25-staff-1
Communication · 20 min · Soft skills + weekly review: Free choice: your weakest soft skill → Week 25 drills · day 7 w25-comm-7
Review · 40 min · Retro of weeks 1-24: what stuck, what didn't; update skip list and next-6-month goals → Decision-making under ambiguity w25-rev-1