Skip to main content
Learn · Concept Library

Concept Library

Deep-dive concept guides with frameworks, code examples, and interview questions. Pick a track — each guide teaches you how to think, not just what to memorize.

276
guides
211
hours

53

guides

Machine Learning

Classical ML and deep learning from first principles through production. Loss functions, regularization, feature engineering, and debugging — theory plus code plus failure modes.

  • Foundations · Classical ML · Deep Learning
  • Stats & Probability for ML interviews
  • Calibration, bootstrap, time series
Browse guides

33

guides

GenAI & Agents

The stack reshaping interviews in 2026: LLMs, RAG, fine-tuning, inference optimization, agent frameworks, and evaluation — built for candidates targeting AI-first roles.

  • Foundations · Generation · RAG
  • Serving, inference, fine-tuning
  • Agents, tool use, evaluation
Browse guides

44

guides

ML System Design

Recommendation, ranking, search, fraud detection — end-to-end ML systems with serving architectures, feature stores, training pipelines, and online/offline evaluation.

  • Building blocks: feature stores, vector DBs, model serving
  • Canonical framework for any MLSD round
  • News Feed ranking · Fraud · Ads ranking
Browse guides

22

guides

Data Structures & Algorithms

Pattern-driven problem solving. Each guide teaches a transferable pattern with complexity analysis and edge-case traps — not another list of 500 random problems.

  • Two Pointers · Sliding Window · Monotonic Stack
  • Trees, Graphs, Dynamic Programming
  • Sorting, Search, Greedy
Browse guides

42

guides

High-Level Design

Distributed systems the way FAANG asks them: latency budgets, capacity math, sharding strategies, and failure modes — grounded in real system blueprints.

  • Consistent Hashing · Sharding · Replication
  • Caching, Queues, Rate Limiting
  • News Feed · URL Shortener · Chat
Browse guides

27

guides

Low-Level Design

SOLID, design patterns, and object-oriented design — interview-grade walkthroughs of Parking Lot, Chess, Splitwise, and more, built around how real engineers actually design.

  • SOLID principles + 5 framework steps
  • Creational · Structural · Behavioral patterns
  • End-to-end case studies
Browse guides

20

guides

Production Engineering

The reactive layer: how senior engineers respond when systems are already running. Metric anomaly triage, incident command, A/B experiment interpretation, and judgment calls about running systems — the hidden curriculum of staff-level on-call. Distinct from Engineering Craft, which covers proactive building and leading.

  • Debugging: metric anomalies, data quality, instrumentation failures
  • Incidents & on-call: escalation, RCA, postmortem culture
  • Experimentation & trade-offs in live systems
Browse guides

13

guides

Engineering Craft

The proactive breadth layer for ML, SDE, and AI engineers: product sense, DevOps and platform fundamentals, data engineering patterns, and staff-level technical leadership. Not debugging production — designing better systems, making better decisions, and growing your scope of impact.

  • Product thinking: north star metrics, driver trees, product sense
  • DevOps & platform: CI/CD, observability stack design, SLOs
  • Technical leadership: estimation, design docs, staff+ arc
Browse guides

22

guides

SQL, Analytics & Scenarios

The fluency layer tested across data, PM, and analyst rounds: SQL joins and window functions, product metrics frameworks, funnel analysis, and open-ended scenario frameworks for staff-level judgment questions.

  • SQL: joins, GROUP BY, CTEs, window functions, indexes
  • Analytics: funnel analysis, north-star metrics, A/B interpretation
  • Scenarios: ambiguous problems, trade-offs, estimation rounds
Browse guides