Tools
Interactive visualizers and technical guides.
Visualizers
Rectangle Plotter
Geometry visualizer
Type a list of rectangles and inspect their overlaps, coordinates, and area.
Graph Visualizer (opens in a new tab)
Tree and graph visualizer
Paste an edge list and drag a node to reroot the tree.
Guides and practice
Rate Limiting with Redis
Interactive guide
Compare four rate-limiting algorithms, inspect their Redis commands, and try the sliding-log simulator.
C++ Prep
Five guides
Fundamentals, virtual functions, iterators, move semantics, and performance.
System Design Prep (opens in a new tab)
Flashcards and rehearsal boards
Spaced-repetition flashcards across twelve system-design topics. Progress stays in your browser.
Cocodessert (opens in a new tab)
AI engineering interview prep
LLM internals, applied engineering, and evaluations.
System design guides
Infrastructure references and worked sizing examples.
Kafka for System Design (opens in a new tab)
Reference board
Kafka explained from the append-only log: core objects, keys and partitioning, delivery semantics, interview patterns, and operations.
Kafka Sizing (opens in a new tab)
Worked example
Size a two-topic deployment from traffic to brokers, partitions, consumers, retention, and cost.
Redis for System Design (opens in a new tab)
Reference board
Data structures, atomicity tools, interview patterns, replication and cluster, and the traps worth naming out loud.
Cassandra at 1M Writes per Second (opens in a new tab)
Worked example
Sizing, setup, keys, tokens, replication, and routing for a messenger writing one million messages a second.
DynamoDB at 1M Writes per Second (opens in a new tab)
Worked example
The same messenger workload on DynamoDB: capacity units, key design, warm throughput, and hot partitions.
Geo Services (opens in a new tab)
Field guide
Nearby search, spatial indexes, moving objects, and sharding, with worked designs for ride dispatch, dating, local search, and delivery.
SQL Locking (opens in a new tab)
Cheat sheet
When you need row locks and when a single statement is enough. Postgres first, with MySQL differences at the end.