Skip to content

Gallery

40 reference diagrams.

Each one is a spec in the repository, rendered here by the same CLI you install. Open one for its YAML and every output format; orrery init <name> copies it as a starting point. Every picture starts in the stage skin, an animated scene in one tint; switch to classic for each vendor's own style.

AI coding agent on the Model Context Protocol: a developer prompts the agent loop in the MCP host, which asks Claude and calls tools through three MCP clients over stdio and Streamable HTTPAI coding agent on the Model Context Protocol: a developer prompts the agent loop in the MCP host, which asks Claude and calls tools through three MCP clients over stdio and Streamable HTTPAI coding agent on the Model Context Protocol: a developer prompts the agent loop in the MCP host, which asks Claude and calls tools through three MCP clients over stdio and Streamable HTTPAI coding agent on the Model Context Protocol: a developer prompts the agent loop in the MCP host, which asks Claude and calls tools through three MCP clients over stdio and Streamable HTTP

AI coding agent on the Model Context Protocol

an AI coding agent on the Model Context Protocol: the MCP host runs one MCP client per server, stdio to a local filesystem server, Streamable HTTP with OAuth to remote GitHub and Postgres servers, Claude choosing each tool, and OTLP traces to Grafana Cloud.

logostabler15 nodes
Data lake on AWS with raw, curated and analytics zonesData lake on AWS with raw, curated and analytics zonesData lake on AWS with raw, curated and analytics zonesData lake on AWS with raw, curated and analytics zones

Data lake on AWS

Firehose and DMS land data in raw, curated and analytics zones refined by Glue jobs, with the catalog and Lake Formation beside the zones and Athena, Redshift and Quick Sight at the bottom.

awslogos15 nodes
Event-driven order processing on AWS with numbered stepsEvent-driven order processing on AWS with numbered stepsEvent-driven order processing on AWS with numbered stepsEvent-driven order processing on AWS with numbered steps

Event-driven order processing on AWS

an EventBridge bus fans orders out to a Step Functions workflow, an SQS billing queue and an SNS topic, with numbered `step:` edges on the main path.

aws12 nodes
Retrieval-augmented generation on AWS with Bedrock and OpenSearch ServerlessRetrieval-augmented generation on AWS with Bedrock and OpenSearch ServerlessRetrieval-augmented generation on AWS with Bedrock and OpenSearch ServerlessRetrieval-augmented generation on AWS with Bedrock and OpenSearch Serverless

Retrieval augmented generation on AWS

an ingestion lane (S3, Lambda) and a query lane (Cognito, API Gateway, a Bedrock AgentCore agent) meet at shared services: a Titan Text Embeddings V2 model that embeds both the chunks and the question, a Bedrock text model, and an OpenSearch Serverless vector index.

aws10 nodes
Hybrid multi-Region AWS landing zone: a corporate data center on Direct Connect with VPN backup, a management account with Control Tower, Organizations and IAM Identity Center, Security and Infrastructure OUs with logging, threat detection, peered Transit Gateways and Network Firewall, and workloads in us-east-1 and us-west-2Hybrid multi-Region AWS landing zone: a corporate data center on Direct Connect with VPN backup, a management account with Control Tower, Organizations and IAM Identity Center, Security and Infrastructure OUs with logging, threat detection, peered Transit Gateways and Network Firewall, and workloads in us-east-1 and us-west-2Hybrid multi-Region AWS landing zone: a corporate data center on Direct Connect with VPN backup, a management account with Control Tower, Organizations and IAM Identity Center, Security and Infrastructure OUs with logging, threat detection, peered Transit Gateways and Network Firewall, and workloads in us-east-1 and us-west-2Hybrid multi-Region AWS landing zone: a corporate data center on Direct Connect with VPN backup, a management account with Control Tower, Organizations and IAM Identity Center, Security and Infrastructure OUs with logging, threat detection, peered Transit Gateways and Network Firewall, and workloads in us-east-1 and us-west-2

Hybrid multi-Region landing zone on AWS

an AWS landing zone with AWS Organizations OUs for security, infrastructure and workloads, IAM Identity Center federated to a corporate IdP, central logging and threat detection, Network Firewall inspection, and a data center on Direct Connect with VPN backup reaching two Regions through peered Transit Gateways.

aws24 nodes
Hybrid network on AWS with Direct Connect, Site-to-Site VPN and Transit GatewayHybrid network on AWS with Direct Connect, Site-to-Site VPN and Transit GatewayHybrid network on AWS with Direct Connect, Site-to-Site VPN and Transit GatewayHybrid network on AWS with Direct Connect, Site-to-Site VPN and Transit Gateway

Hybrid network on AWS

a corporate data center reaches VPCs in three AWS accounts in us-east-1 through a Transit Gateway, over Direct Connect with a Site-to-Site VPN backup.

aws12 nodes
Serverless web API on AWS with CloudFront, S3, API Gateway, Lambda, DynamoDB and CognitoServerless web API on AWS with CloudFront, S3, API Gateway, Lambda, DynamoDB and CognitoServerless web API on AWS with CloudFront, S3, API Gateway, Lambda, DynamoDB and CognitoServerless web API on AWS with CloudFront, S3, API Gateway, Lambda, DynamoDB and Cognito

Serverless web API on AWS

CloudFront (a global service, so drawn outside the Region) serves a single-page app from S3 and sends `/api/*` to an API Gateway REST API, Lambda and DynamoDB, with a Cognito user pool authorizer; the function also writes to a receipts bucket.

aws8 nodes
The three-tier web application on AWS in the dark themeThe three-tier web application on AWS in the dark themeThe three-tier web application on AWS in the dark themeThe three-tier web application on AWS in the dark theme

Three-tier web application on AWS

the same diagram with `theme: dark`.

aws11 nodes
Three-tier web application on AWS across two Availability ZonesThree-tier web application on AWS across two Availability ZonesThree-tier web application on AWS across two Availability ZonesThree-tier web application on AWS across two Availability Zones

Three-tier web application on AWS

a VPC across two Availability Zones with public, web and data subnets, NAT gateways, an ALB, and one Auto Scaling group drawn across both zones with `around:`; zone B writes to the primary database in zone A.

aws11 nodes
AKS application platform on AzureAKS application platform on AzureAKS application platform on AzureAKS application platform on Azure

AKS application platform on Azure

Application Gateway in front of an AKS cluster with system and user node pools, Container Registry, workload identity (the pods exchange their token with Microsoft Entra ID, then read Key Vault), and Log Analytics.

azure8 nodes
Hub-spoke network on Azure with ExpressRoute, a VPN backup, Azure Firewall, Bastion and two peered spokesHub-spoke network on Azure with ExpressRoute, a VPN backup, Azure Firewall, Bastion and two peered spokesHub-spoke network on Azure with ExpressRoute, a VPN backup, Azure Firewall, Bastion and two peered spokesHub-spoke network on Azure with ExpressRoute, a VPN backup, Azure Firewall, Bastion and two peered spokes

Hub-spoke network on Azure

a hub-spoke network on Azure: on-premises reaches the hub's ExpressRoute and VPN gateways and Azure Firewall, which forwards over the peerings to the spokes, with Bastion in the hub, AKS in one spoke and VMs with a private endpoint to Azure SQL in the other, Private DNS and Log Analytics.

azure13 nodes
Retrieval augmented generation on Azure: Application Gateway, Container Apps, private endpoints, Azure OpenAI, Azure AI Search and Blob Storage ingestionRetrieval augmented generation on Azure: Application Gateway, Container Apps, private endpoints, Azure OpenAI, Azure AI Search and Blob Storage ingestionRetrieval augmented generation on Azure: Application Gateway, Container Apps, private endpoints, Azure OpenAI, Azure AI Search and Blob Storage ingestionRetrieval augmented generation on Azure: Application Gateway, Container Apps, private endpoints, Azure OpenAI, Azure AI Search and Blob Storage ingestion

Retrieval augmented generation on Azure

Azure OpenAI chat grounded in an Azure AI Search index, behind Application Gateway and private endpoints, with an indexer filling the index from Blob Storage.

azure14 nodes
Web app on Azure with Front Door, App Service and private endpointsWeb app on Azure with Front Door, App Service and private endpointsWeb app on Azure with Front Door, App Service and private endpointsWeb app on Azure with Front Door, App Service and private endpoints

Web app on Azure

Front Door to App Service outside the VNet, with only its integration interface in the app subnet, reaching Azure SQL and Key Vault through private endpoints and Private DNS zones, with Azure's filled arrowheads.

azure10 nodes
C4 container diagram for an online storeC4 container diagram for an online storeC4 container diagram for an online storeC4 container diagram for an online store

Container diagram for an online store

a C4 container view built from shapes: two people, a `c4:boundary` system, its containers and database, and grey external systems.

c48 nodes
Open-source data platform with Debezium, Airbyte, Kafka, Flink, Iceberg on Ceph, Hive Metastore, dbt, Trino and SupersetOpen-source data platform with Debezium, Airbyte, Kafka, Flink, Iceberg on Ceph, Hive Metastore, dbt, Trino and SupersetOpen-source data platform with Debezium, Airbyte, Kafka, Flink, Iceberg on Ceph, Hive Metastore, dbt, Trino and SupersetOpen-source data platform with Debezium, Airbyte, Kafka, Flink, Iceberg on Ceph, Hive Metastore, dbt, Trino and Superset

Open-source data platform

an open-source lakehouse: Debezium and Airbyte feed Kafka, Flink lands Iceberg tables on Ceph with a Hive Metastore catalog, dbt models them on Trino for Superset, on an Airflow schedule.

logossitabler13 nodes
Order approval flowchart with terminators, an input parallelogram, a predefined credit check, yes/no decisions, manual steps, a delay and a documentOrder approval flowchart with terminators, an input parallelogram, a predefined credit check, yes/no decisions, manual steps, a delay and a documentOrder approval flowchart with terminators, an input parallelogram, a predefined credit check, yes/no decisions, manual steps, a delay and a documentOrder approval flowchart with terminators, an input parallelogram, a predefined credit check, yes/no decisions, manual steps, a delay and a document

Order approval flowchart

an ISO 5807 flowchart (`kind: flowchart`) of order approval. Order details come in through a data parallelogram and a predefined credit check decides; a failed check asks for prepayment (a manual operation) and waits up to 48 hours (a delay), an order over the limit needs a manager's sign-off, and approved orders end with a printed invoice (a document).

13 nodes
GKE application platform on Google Cloud with Cloud Build, Artifact Registry, Cloud Deploy, Memorystore, Secret Manager and Cloud SQLGKE application platform on Google Cloud with Cloud Build, Artifact Registry, Cloud Deploy, Memorystore, Secret Manager and Cloud SQLGKE application platform on Google Cloud with Cloud Build, Artifact Registry, Cloud Deploy, Memorystore, Secret Manager and Cloud SQLGKE application platform on Google Cloud with Cloud Build, Artifact Registry, Cloud Deploy, Memorystore, Secret Manager and Cloud SQL

GKE application platform on Google Cloud

a GKE platform in Google Cloud's frames (cloud, project, region, cluster): Cloud Build pushes each image to Artifact Registry and releases it through Cloud Deploy to two namespaces, next to Memorystore, Secret Manager and Cloud SQL, with Cloud Logging and Cloud Monitoring.

gcpk8slogos12 nodes
Streaming analytics on Google Cloud with Pub/Sub, Dataflow and BigQueryStreaming analytics on Google Cloud with Pub/Sub, Dataflow and BigQueryStreaming analytics on Google Cloud with Pub/Sub, Dataflow and BigQueryStreaming analytics on Google Cloud with Pub/Sub, Dataflow and BigQuery

Streaming analytics on Google Cloud

Pub/Sub and Dataflow stream events into BigQuery and Cloud Storage and score them with a Vertex AI model, with Looker dashboards on BigQuery.

gcp8 nodes
MLOps on Google Cloud: Cloud Build ships pipeline code to Agent Platform Pipelines, which trains, registers and serves a model; drift alerts restart it through Cloud Logging, Pub/Sub and Cloud Run functionsMLOps on Google Cloud: Cloud Build ships pipeline code to Agent Platform Pipelines, which trains, registers and serves a model; drift alerts restart it through Cloud Logging, Pub/Sub and Cloud Run functionsMLOps on Google Cloud: Cloud Build ships pipeline code to Agent Platform Pipelines, which trains, registers and serves a model; drift alerts restart it through Cloud Logging, Pub/Sub and Cloud Run functionsMLOps on Google Cloud: Cloud Build ships pipeline code to Agent Platform Pipelines, which trains, registers and serves a model; drift alerts restart it through Cloud Logging, Pub/Sub and Cloud Run functions

MLOps on Google Cloud with Agent Platform

MLOps on Gemini Enterprise Agent Platform (formerly Vertex AI): Cloud Build ships the pipeline, Pub/Sub and Cloud Run functions start it, Dataflow and Managed Training feed Model Registry and an online endpoint, and drift alerts retrain.

gcplogos18 nodes
Web application on Google Cloud drawn as product cardsWeb application on Google Cloud drawn as product cardsWeb application on Google Cloud drawn as product cardsWeb application on Google Cloud drawn as product cards

Web application on Google Cloud

Google Cloud product cards (`look: card`), a global load balancer in front of Cloud Run with Cloud CDN and Cloud Armor attached to its backend, and Cloud Run reaching Cloud SQL, Secret Manager and Memorystore (in a VPC network, over Direct VPC egress).

gcp8 nodes
CI/CD from GitHub Actions to AWS with OIDC, ECR, CloudFormation and ECSCI/CD from GitHub Actions to AWS with OIDC, ECR, CloudFormation and ECSCI/CD from GitHub Actions to AWS with OIDC, ECR, CloudFormation and ECSCI/CD from GitHub Actions to AWS with OIDC, ECR, CloudFormation and ECS

CI/CD from GitHub Actions to AWS

a merge to main runs GitHub Actions, which assumes an IAM role with its OIDC token, then pushes to ECR and deploys to ECS with CloudFormation.

awsgithublogos7 nodes
GitOps on Kubernetes: GitHub Actions, ghcr.io, a config repository and Argo CD syncing staging and production clustersGitOps on Kubernetes: GitHub Actions, ghcr.io, a config repository and Argo CD syncing staging and production clustersGitOps on Kubernetes: GitHub Actions, ghcr.io, a config repository and Argo CD syncing staging and production clustersGitOps on Kubernetes: GitHub Actions, ghcr.io, a config repository and Argo CD syncing staging and production clusters

GitOps delivery with Argo CD

GitOps with Argo CD: a pull request runs GitHub Actions, which pushes the image to ghcr.io and bumps its tag in the config repository; Argo CD syncs the staging cluster automatically and production on a manual sync, with Prometheus and Grafana monitoring production.

githubk8slogos16 nodes
Web application on Red Hat OpenShift on IBM Cloud across three zones in a VPCWeb application on Red Hat OpenShift on IBM Cloud across three zones in a VPCWeb application on Red Hat OpenShift on IBM Cloud across three zones in a VPCWeb application on Red Hat OpenShift on IBM Cloud across three zones in a VPC

Web application on Red Hat OpenShift on IBM Cloud

IBM Cloud Internet Services in front of a static site in Cloud Object Storage and a three-zone Red Hat OpenShift cluster, with App ID and services reached through Virtual Private Endpoints.

ibm20 nodes
SaaS onboarding customer journey map: five stages, four lanes of step cards and an emotion curve across the stagesSaaS onboarding customer journey map: five stages, four lanes of step cards and an emotion curve across the stagesSaaS onboarding customer journey map: five stages, four lanes of step cards and an emotion curve across the stagesSaaS onboarding customer journey map: five stages, four lanes of step cards and an emotion curve across the stages

SaaS onboarding journey

a customer journey map (`kind: journey`): five stages from Discover to Renew as columns, lanes for customer actions, touchpoints, pain points and opportunities as rows, the customer's path as arrows, and an emotion curve from each step's score that dips at the data import and peaks at the first shared report.

17 nodes
Kubernetes web workload with the official Kubernetes iconsKubernetes web workload with the official Kubernetes iconsKubernetes web workload with the official Kubernetes iconsKubernetes web workload with the official Kubernetes icons

Kubernetes web workload

the official Kubernetes icons inside a namespace: Ingress, Service, Deployment, ReplicaSet, Pods, HorizontalPodAutoscaler, ConfigMap, Secret and a PersistentVolumeClaim bound to a PersistentVolume.

k8s11 nodes
Order desk flowchart: customer order, order desk, credit check, a within-limit decision with a sales manager branch, pick and pack, then courier to delivered and invoice to paidOrder desk flowchart: customer order, order desk, credit check, a within-limit decision with a sales manager branch, pick and pack, then courier to delivered and invoice to paidOrder desk flowchart: customer order, order desk, credit check, a within-limit decision with a sales manager branch, pick and pack, then courier to delivered and invoice to paidOrder desk flowchart: customer order, order desk, credit check, a within-limit decision with a sales manager branch, pick and pack, then courier to delivered and invoice to paid

Order desk

a business flowchart from a customer's order to a delivered parcel and a paid invoice. The order desk, the credit check and the packing bench are the team's own SVG icons (`local:`), drawn to the IC01 to IC10 rules and passed by `orrery icons check`; the other steps use the business `generic:` names.

local11 nodes
Mind map: Product launch in the middle, with Market, Sales and Support on the right and Product and Comms on the left, each with its ideasMind map: Product launch in the middle, with Market, Sales and Support on the right and Product and Comms on the left, each with its ideasMind map: Product launch in the middle, with Market, Sales and Support on the right and Product and Comms on the left, each with its ideasMind map: Product launch in the middle, with Market, Sales and Support on the right and Product and Comms on the left, each with its ideas

Product launch

a product launch as a two-sided mind map (`kind: mindmap`): five workstreams split between the halves with fifteen ideas, curved branches thick out of the root and one color per workstream.

21 nodes
ML platform on AWS with SageMaker AI and Step FunctionsML platform on AWS with SageMaker AI and Step FunctionsML platform on AWS with SageMaker AI and Step FunctionsML platform on AWS with SageMaker AI and Step Functions

ML platform on AWS

SageMaker AI Processing into Feature Store, a Step Functions training pipeline, Model Registry, batch and real-time serving, and a drift alarm that starts retraining.

aws11 nodes
A modern SaaS stack: Cloudflare, Next.js on Vercel, Neon, Upstash, Stripe, Clerk, Resend, Sentry and PostHogA modern SaaS stack: Cloudflare, Next.js on Vercel, Neon, Upstash, Stripe, Clerk, Resend, Sentry and PostHogA modern SaaS stack: Cloudflare, Next.js on Vercel, Neon, Upstash, Stripe, Clerk, Resend, Sentry and PostHogA modern SaaS stack: Cloudflare, Next.js on Vercel, Neon, Upstash, Stripe, Clerk, Resend, Sentry and PostHog

Modern SaaS web stack

a SaaS web stack drawn as cards: Next.js on Vercel serving the UI, Cloudflare in front of the Node.js API, Neon and Upstash for data, Stripe, Resend and Clerk as third-party services, Sentry and PostHog for observability.

logossi12 nodes
AWS, Google Cloud and Azure with a shared identity providerAWS, Google Cloud and Azure with a shared identity providerAWS, Google Cloud and Azure with a shared identity providerAWS, Google Cloud and Azure with a shared identity provider

Multi-cloud with a shared identity provider

AWS, Google Cloud and Azure in their own vendor frames (the Azure subscription inside its Microsoft Entra tenant), federated to one Okta tenant and joined at a colocation router by Direct Connect, Partner Interconnect and ExpressRoute, each reaching its workload through a gateway or router in a VPC or VNet.

awsazuregcplogos12 nodes
Observability stack with OpenTelemetry, Prometheus, Loki, Tempo and GrafanaObservability stack with OpenTelemetry, Prometheus, Loki, Tempo and GrafanaObservability stack with OpenTelemetry, Prometheus, Loki, Tempo and GrafanaObservability stack with OpenTelemetry, Prometheus, Loki, Tempo and Grafana

Observability stack with OpenTelemetry

OpenTelemetry SDKs export to a Collector that fans out to Prometheus, Loki and Tempo, with Grafana querying them and Alertmanager routing to Slack and paging PagerDuty.

logos11 nodes
BPMN order-to-cash process: a customer pool and a supplier pool with sales, warehouse and finance lanes, exchanging purchase order, invoice, payment and goods messagesBPMN order-to-cash process: a customer pool and a supplier pool with sales, warehouse and finance lanes, exchanging purchase order, invoice, payment and goods messagesBPMN order-to-cash process: a customer pool and a supplier pool with sales, warehouse and finance lanes, exchanging purchase order, invoice, payment and goods messagesBPMN order-to-cash process: a customer pool and a supplier pool with sales, warehouse and finance lanes, exchanging purchase order, invoice, payment and goods messages

Order to cash

order to cash in BPMN 2.0 (`kind: process`): a Customer pool and a Supplier pool with Sales, Warehouse and Finance lanes, run top to bottom. A message start event, user, send, receive and service tasks, an exclusive gateway with Yes and No branches, a parallel split and join, and four message flows between the pools.

16 nodes
Retrieval-augmented generation with Qdrant, bge-m3, LangChain, FastAPI, Ollama and LangfuseRetrieval-augmented generation with Qdrant, bge-m3, LangChain, FastAPI, Ollama and LangfuseRetrieval-augmented generation with Qdrant, bge-m3, LangChain, FastAPI, Ollama and LangfuseRetrieval-augmented generation with Qdrant, bge-m3, LangChain, FastAPI, Ollama and Langfuse

Retrieval-augmented generation with open components

RAG with open components: an ingestion lane chunks and embeds documents into Qdrant, and a React, FastAPI and LangChain query lane answers with a local Ollama model or a hosted one, traced in Langfuse.

logossitabler14 nodes
Web app on AWS: users, an Application Load Balancer and an ECS service with a database and a cache in a VPC, and an S3 bucket in the RegionWeb app on AWS: users, an Application Load Balancer and an ECS service with a database and a cache in a VPC, and an S3 bucket in the RegionWeb app on AWS: users, an Application Load Balancer and an ECS service with a database and a cache in a VPC, and an S3 bucket in the RegionWeb app on AWS: users, an Application Load Balancer and an ECS service with a database and a cache in a VPC, and an S3 bucket in the Region

Web app on AWS

the `orrery init` default, a load-balanced container service with its database and cache in a VPC and a bucket in the Region, inside one AWS Cloud group.

aws6 nodes
State machine: an order goes from Created to Paid, Packed, Shipped and Delivered, can be Cancelled or Refunded, and loops on Created when editedState machine: an order goes from Created to Paid, Packed, Shipped and Delivered, can be Cancelled or Refunded, and loops on Created when editedState machine: an order goes from Created to Paid, Packed, Shipped and Delivered, can be Cancelled or Refunded, and loops on Created when editedState machine: an order goes from Created to Paid, Packed, Shipped and Delivered, can be Cancelled or Refunded, and loops on Created when edited

Order lifecycle

an order's lifecycle as a UML state machine (`kind: state`): initial and final states, Created through Delivered with Cancelled and Refunded, transitions written event [guard] / action, and a self-transition looped on the corner of Created.

10 nodes
Swimlane flowchart of incident response across the on-call engineer, incident lead and communications lanes, from the alert to the postmortemSwimlane flowchart of incident response across the on-call engineer, incident lead and communications lanes, from the alert to the postmortemSwimlane flowchart of incident response across the on-call engineer, incident lead and communications lanes, from the alert to the postmortemSwimlane flowchart of incident response across the on-call engineer, incident lead and communications lanes, from the alert to the postmortem

Incident response

a cross-functional flowchart of incident response in three standalone lanes (On-call engineer, Incident lead, Communications), with terminators, decisions and a mitigation loop.

13 nodes
Chat system design drawn with shapesChat system design drawn with shapesChat system design drawn with shapesChat system design drawn with shapes

Chat system design

shape nodes for a whiteboard look (cylinders for stores, a queue for the Kafka topic) around WebSocket gateways, chat, presence and push services, media storage and a CDN.

logossitabler13 nodes
URL shortener system design with neutral icons and product logosURL shortener system design with neutral icons and product logosURL shortener system design with neutral icons and product logosURL shortener system design with neutral icons and product logos

URL shortener system design

vendor-neutral `generic:` icons and product logos, from CDN, load balancer and API servers to Redis, PostgreSQL and a Kafka, Flink and ClickHouse click pipeline.

logossitabler11 nodes
Video streaming system design with upload, transcoding, CDN playback, metadata, search and recommendationsVideo streaming system design with upload, transcoding, CDN playback, metadata, search and recommendationsVideo streaming system design with upload, transcoding, CDN playback, metadata, search and recommendationsVideo streaming system design with upload, transcoding, CDN playback, metadata, search and recommendations

Video streaming system design

uploads go to object storage through a pre-signed URL, a queue feeds transcoding workers that write HLS and DASH renditions, and viewers stream segments from the CDN while view events flow through Kafka to recommendations.

logostabler14 nodes
Org chart: the Chief Product Officer over Design, Engineering and Product management, each over its teamsOrg chart: the Chief Product Officer over Design, Engineering and Product management, each over its teamsOrg chart: the Chief Product Officer over Design, Engineering and Product management, each over its teamsOrg chart: the Chief Product Officer over Design, Engineering and Product management, each over its teams

Product organization

a product organization as an org chart (`kind: tree`): the chief product officer, three directors and their eight teams as cards with generic icons, each parent centered over its reports on a shared connector bar.

11 nodes