diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
new file mode 100644
index 0000000..368e2f2
--- /dev/null
+++ b/CONTRIBUTING.md
@@ -0,0 +1,93 @@
+## Development
+
+#### Building from Source
+
+```bash
+# Standard build (executor image pulled at runtime)
+docker build -t code-interpreter -f code-interpreter/Dockerfile .
+
+# Build with pre-loaded executor image for instant DinD startup
+./build-preloaded.sh code-interpreter:preloaded
+```
+
+**Build arguments:**
+- `SKIP_NESTED_DOCKER=1` - Skip installing Docker entirely (only for Docker-out-of-Docker mode)
+- `PYTHON_EXECUTOR_DOCKER_IMAGE=custom/image` - Use a custom executor image
+
+#### Pre-loaded Images for Faster Startup
+
+For production or offline environments, you can build an image with the executor pre-embedded:
+
+```bash
+# Build pre-loaded image (includes executor, ~1GB larger but instant DinD startup)
+./build-preloaded.sh code-interpreter:preloaded
+
+# Run with instant startup (no pulling needed)
+docker run --rm -it \
+  --privileged \
+  -p 8000:8000 \
+  code-interpreter:preloaded
+```
+
+This is ideal for:
+- Production deployments (predictable startup times)
+- Air-gapped/offline environments
+- CI/CD pipelines
+- Any scenario where you want instant DinD readiness
+
+### Local Deployment
+
+#### Prerequisites
+
+- Python 3.11
+- Docker (for execution isolation)
+- [uv](https://github.com/astral-sh/uv) package manager (recommended)
+
+#### Running the Service
+
+```bash
+# Clone the repository
+git clone https://github.com/onyx-dot-app/code-interpreter.git
+cd code-interpreter
+
+# Activate a virtual environment
+python -m venv .venv
+source .venv/bin/activate
+
+# Install dependencies
+uv sync --locked
+```
+
+```bash
+# Start the API server (defaults to 0.0.0.0:8000)
+code-interpreter-api
+
+# Or specify host/port via environment variables
+HOST=127.0.0.1 PORT=8080 code-interpreter-api
+```
+
+### Code Quality
+
+```bash
+# Type checking (must pass strict mypy)
+mypy .
+
+# Linting
+ruff check .
+
+# Formatting
+ruff format .
+
+# Run all pre-commit hooks
+pre-commit run --all-files
+```
+
+### Testing
+
+```bash
+# Run integration tests
+pytest tests/integration_tests -q
+
+# Run a specific test
+pytest tests/integration_tests/test_file.py::test_function_name -v
+```
diff --git a/README.md b/README.md
index f001f7b..fed0b62 100644
--- a/README.md
+++ b/README.md
@@ -6,9 +6,11 @@ Everything runs locally and the execution environment comes pre-packaged with a
 
 This project aims to be the easiest, lightest weight way to add secure Python execution to your AI agent.
 
+![Code Interpreter](code_interpreter_exp.png)
+
 ## How it works
 
-The security first architecture and an overview of the implementation can be found [here](https://github.com/onyx-dot-app/code-interpreter/blob/main/HOW_IT_WORKS.md).
+The security first architecture and an overview of the implementation can be found [here](HOW_IT_WORKS.md).
 
 ## Quick Start
 
@@ -61,7 +63,7 @@ docker run --rm -it \
 
 ### Kubernetes Deployment
 
-See [here](https://github.com/onyx-dot-app/code-interpreter/blob/main/kubernetes/code-interpreter/README.md) for Helm and K8s deployment instructions
+See [here](kubernetes/code-interpreter/README.md) for Helm and K8s deployment instructions
 
 ## API Usage
 
diff --git a/code_interpreter_exp.png b/code_interpreter_exp.png
new file mode 100644
index 0000000..04469fb
Binary files /dev/null and b/code_interpreter_exp.png differ