diff --git a/README.md b/README.md
index 21b97cd..52f8c3b 100644
--- a/README.md
+++ b/README.md
@@ -4,6 +4,22 @@ The core TTS models that power FakeYou.
 
 I'm in the progress of pulling these out of the service monorepo so we can iterate faster.
 
+How This Works
+--------------
+We have a Rust monolith that controls the user interface, account system, and all the database CRUD operations.
+It doesn't do any ML work itself or have any attached GPUs.
+
+There are a series of worker pods ("jobs") that pull from work queues and run inference, then upload the results.
+
+![](docs/diagrams/microservices.png)
+
+The jobs are written in Rust and either shell out to Python code or call an in-container Python server that attempts 
+to LRU cache models in memory.
+
+![](docs/diagrams/jobs1.png)
+
+![](docs/diagrams/jobs2.png)
+
 TODO
 ----
 
diff --git a/docs/diagrams/jobs1.png b/docs/diagrams/jobs1.png
new file mode 100644
index 0000000..082c1c8
Binary files /dev/null and b/docs/diagrams/jobs1.png differ
diff --git a/docs/diagrams/jobs2.png b/docs/diagrams/jobs2.png
new file mode 100644
index 0000000..f437b35
Binary files /dev/null and b/docs/diagrams/jobs2.png differ
diff --git a/docs/diagrams/microservices.png b/docs/diagrams/microservices.png
new file mode 100644
index 0000000..fb05d28
Binary files /dev/null and b/docs/diagrams/microservices.png differ