How modern software is written, shipped and kept running

A handbook for software engineers: the ideas, the products that implement them, and real stacks that put them together.

Try:

Learn a conceptWhat a queue, a lakehouse or an event loop is, how it works, and when to reach for it. 133 concepts, each with a live model.Message queues Lakehouses Event loop
Look up a technologyWhat a product is, why teams pick it, what to watch out for, and what it works with. 405 technologies, every fact sourced.Apache Kafka PostgreSQL DuckDB
Choose between optionsIdeas and products that are easy to mix up, side by side, with a rule of thumb. Every concept page also compares its technologies in a table.Object-oriented, functional or service layer?Static or dynamic dispatch?Event loop or threads?

Or watch 10 real stacks run live, or design your own in the playground.

The whole system, at concept level

Click any box to learn the concept.

Live modeldata or requestresponsebusy / idle instance

What you’re watchingTwo journeys through one system. On top, sensor readings flow through a broker into a stream, where functions write them to a time-series database while an archiver batches them into object storage. Below, an operator’s dashboard request passes a CDN, a gateway and API pods, and is answered from cache when it can. Click any underlined part to open it.

Browse by area

BuildHow code is written, and how a system is split into parts.Programming 23 System design 6 Developer tools 7
InfrastructureWhere code runs, and how its parts reach each other.Compute 11 Networking 11 Messaging 7 Storage 4
DataStoring, shaping and querying data.Databases 14 Data modeling 8 Analytics 11
Ship & runGetting changes to production, then keeping it healthy and safe.Delivery 10 Observability 11 Security 6
TeamsWorking together: how teams are shaped and how work flows.Organization 4

Real stacks

Products wired together, running live, explained step by step.

Small lakehouse on DuckLakeDuckDB writes and queries tables whose catalog lives in PostgreSQL and whose data sits as Parquet on S3.DuckDBDuckLakePostgreSQLAmazon S3 Open lakehouse on IcebergKafka streams events into Spark, which writes Iceberg tables on S3 that Trino queries for dashboards.Apache KafkaApache SparkApache IcebergAmazon S3Trino Serverless web app on AWSCloudFront serves the site from S3 and sends API calls through API Gateway to Lambda functions backed by DynamoDB.Amazon CloudFrontAmazon S3Amazon API GatewayAWS LambdaAmazon DynamoDB

All 10 architectures →

Commonly confused

All 19 comparisons →