Skip to main content

Kaleido Platform architecture

Kaleido is a platform for casino games with two products that share one code base and one deployment:

  • Kaleido Taxonomy tags games against a hierarchical taxonomy (theme, mechanics, features and so on). Tags come from people, from an in-app AI tagging agent, or from an external AI client over MCP. A person reviews every AI suggestion before it counts as confirmed.
  • Kaleido GRE (the game recommendation engine) turns customer play into per-game affinities and blends them with tag similarity, co-play, popularity and newness into a ranked, explained list of games per customer.

This site describes how the platform is built: the services, what each one owns, how they talk to each other, and how each capability works step by step. Every page is written from the source tree; file paths point at the code that implements the behaviour.

How to read this site​

SectionWhat it answers
System contextWho uses the platform and which outside systems it touches
ContainersEvery deployable unit, its port and its dependencies
Data ownershipWhich service owns which Postgres schema or Redis key
SecurityJWT for people, a shared service key between services, the gateway in between
Request flowsEnd-to-end sequence diagrams of the main journeys
ServicesOne page per service: responsibilities, API, internals, sequences
CapabilitiesOne page per capability: the algorithm or workflow behind it
OperationsDeployment, configuration, observability, failure drills, known limits

The platform in one picture​

Two recommendation paths​

Kaleido GRE has two ways to get from play to recommendations. They share the scoring code and differ in where the customer's affinities come from.

PathData flowFreshnessUsed for
BatchSimulator or platform posts events to Activity (Postgres). Profile rebuilds affinities on demand. Recommender reads them over HTTP.As fresh as the last profile buildTuning, what-if, holdout accuracy tests
LiveEvents go to Kafka, keyed by customer. Ingest keeps decayed per-game totals in Redis. Recommender reads Redis directly (?source=redis).SecondsReal traffic at thousands of events per second

See Live streaming for the live path and Affinity for how both compute "how much does this customer like this game".

Technology​

LayerChoice
ServicesASP.NET Core minimal APIs on .NET 10, EF Core with Npgsql
DatabasePostgreSQL 17 with pgvector (one schema per service)
StreamingApache Kafka (KRaft) with Confluent.Kafka; JSON, Avro or Protobuf payloads
Live stateRedis 7.4 with a Lua script for atomic, idempotent updates
Front endReact 19, TypeScript, Vite, SignalR client
AIAnthropic, OpenAI, OpenAI-compatible endpoints and Cursor agents, configured at runtime
AI interopModel Context Protocol server (streamable HTTP) inside the Taxonomy API
TelemetryOpenTelemetry (traces, metrics, logs) to a collector, then SigNoz
PackagingPodman containers and podman-compose via deploy/deploy.sh