Skip to content

Interviews

Strategy in one paragraph

Interview readiness is a checkpoint, not the goal. Every 4 weeks (weeks 4, 8, 12, 16, 20, 24) you run a mock loop scored with the same unified rubric, so scores are comparable and you can see growth. Each loop is sized to what you have studied so far and ends in a short gap analysis that changes the next 4 weeks of the plan. By week 24 you run a full Staff loop (coding, system design, AI system design, low-level design, behavioral/leadership) and should be scoring at the Staff bar.

Pages

Page Use it for
Unified rubric Scoring every round, Senior vs Staff descriptors, scoring sheet
Checkpoints Exact prompts, targets and remediation per checkpoint
Mock prompt bank 105 prompts across system design, AI system design, LLD, behavioral
AI mock interviewer Copy-paste prompts to run strict mocks with Claude/ChatGPT
Company process How Staff/Principal loops work in 2026 and a prep timeline

Why the same rubric every time

  • Comparable scores: growth only means something if the ruler does not change.
  • Targets rise, rubric doesn't: CP1 expects Senior-bar 3s; CP6 expects Staff-bar 3–4s. The descriptors are fixed; the target column moves.
  • Fast diagnosis: a persistent 2 on one dimension (e.g. "trade-off articulation") is a clearer signal than "SD felt bad".

Loop composition per checkpoint

CP Week Date (week ends) Coding (DSA) System design AI system design LLD Behavioral / Staff Total time
CP1 4 Sun 2026-10-25 2 mediums (45 min) 1 classic (45 min) — (5-min verbal) — 1 story (15 min) ~2 h
CP2 8 Sun 2026-11-22 2 mediums (45 min) 1 storage-heavy (45 min) 1 RAG (45 min) — 2 stories (20 min) ~2.75 h
CP3 12 Sun 2026-12-20 1 medium + 1 hard (60 min) 1 async/streaming (45 min) 1 agent platform (45 min) 1 (45 min) 3 stories (30 min) ~3.75 h
CP4 16 Sun 2027-01-17 2 mediums (45 min) 1 (60 min, Staff scope) 1 gateway/eval (60 min) — Staff leadership round (45 min) ~3.5 h
CP5 20 Sun 2027-02-14 1 medium + 1 hard (60 min) 1 distributed-depth (60 min) 1 serving/cost (60 min) 1 concurrency LLD (45 min) 3 stories (30 min) ~4.25 h
CP6 24 Sun 2027-03-14 2 rounds (2 × 45 min) 1 (60 min) 1 (60 min) 1 (45 min) Staff behavioral + project deep dive (2 × 45 min) ~6.5 h over 2 days

Split long loops across Saturday and Sunday. Details and exact prompts: checkpoints.

How to run mocks

Interviewer options (use at least two kinds per checkpoint)

Option Best for Cost How
Peers / colleagues Behavioral, SD — realistic human pushback Free Trade mocks with a senior/Staff peer; give them the rubric and a prompt from the bank; they score, you swap next week
Paid expert mocks (Hello Interview, interviewing.io, Exponent and similar) Calibrated Staff-bar feedback from current big-tech interviewers Paid per session Book 1 for CP3 or CP4 (calibration) and 1–2 for CP6. Ask the interviewer to score against your rubric in addition to theirs
AI mock interviewer High-frequency reps, any hour, strict time-box ~$0 Use the prompts in AI mock interviewer; voice mode for realism
Recording yourself Communication, filler words, structure, time management Free Record screen + audio (whiteboard tool such as Excalidraw); watch at 1.5× within 24 h; score yourself before reading any feedback
Community mock platforms Coding under a stranger's eye Free–low Peer-matching platforms for DSA; good for CP1–CP3 nerves

Protocol for every mock

  1. Prepare the environment: timer, whiteboard (Excalidraw / tldraw), plain editor without autocomplete for coding, camera on.
  2. No looking things up. If you would not have it in a real interview, you do not have it now.
  3. Time-box strictly. Stop at the bell; unfinished = unfinished.
  4. Score immediately with the scoring sheet: yourself first, then the interviewer/AI; record both.
  5. Write 3 lines: biggest strength, biggest gap, one concrete drill for next week.
  6. Log it in the table below and add any missed concepts as flashcards.

Recording-review checklist

  • Did I restate the problem and clarify requirements in the first 3–5 minutes?
  • Did I say numbers out loud (QPS, storage, latency, cost)?
  • Did I name at least two options before choosing, and say why?
  • How many times did I go silent for more than 20 seconds?
  • Did I drive, or did the interviewer have to pull me forward?
  • Did I finish with risks, bottlenecks and what I would do next?

Score log

Copy this table into docs/log/ or your notes and fill it after each checkpoint. Scores are the round average (1–4) from the scoring sheet.

CP Date Coding SD AI SD LLD Behavioral/Staff Overall Target met? Top gap Plan change
CP0 baseline (wk 0) 2026-09-26/27 — — n/a
CP1 (wk 4) — —
CP2 (wk 8) —
CP3 (wk 12)
CP4 (wk 16) —
CP5 (wk 20)
CP6 (wk 24)

Where the prep material lives