Adds image and SDLC section in overview docs.

docs/concepts/overview.rst

@@ -12,3 +12,30 @@ This is a quick overview of Burr's design -- the concepts are explored in more d
 Note that we did not mention LLMs above at all! That's good -- you want your LLM frameworks to be ever-so-slightly decoupled from them for the best experience (all of AI is just software, plain and simple, after all...).
 
 And that's the basics! Let's dive into the details.
+
+================================
+SDLC with LLM based applications
+================================
+If you're building an LLM-based application, you'll want to follow a slightly different SDLC than you would for a traditional software project. Here's a rough outline of what that might look like:
+
+.. image:: ../_static/burr_sdlc.png
+   :alt: SDLC with LLMs
+   :align: center
+
+The two cycles that exist are:
+
+1. App Dev Loop.
+2. Test Driven Development Loop.
+
+and you will use one to feed into the other, etc.
+
+Walking through the diagram the SDLC looks like this:
+
+1. Write code with Burr.
+2. Use Burr's integrated observability, and trace all parts of your application.
+3. With the data collected, you can: (1) annotate what was captured and export it, or (2) create a pytest fixture with it.
+4. Create a data set from the annotated data or by running tests.
+5. Evaluate the data set.
+6. Analyze the results.
+7. Either adjust code or prompts, or ship the code.
+8. Iterate using one of the loops...

examples/pytest/README.md

@@ -1,3 +1,17 @@
+# A SDLC with Burr and pytest
+Here we show a quick example of a software development lifecycle (SDLC) with Burr and pytest.
+
+![Burr and pytest](burr_sdlc.png)
+
+While we don't cover everything in the diagram, in this example we specifically show how to do most of the TDD loop:
+
+1. Create a test case.
+2. Run the test case.
+3. Create a dataset.
+4. Show how you might construct evaluation logic to evaluate the output of your agent / augmented LLM / application.
+
+using Burr and pytest.
+
 # Using pytest to evaluate your agent  / augmented LLM / application
 
 An agent / augmented LLM is a combination of LLM calls and logic. But how do we know if it's working? Well we can test & evaluate it.