diff --git a/Cargo.toml b/Cargo.toml
index d40d127..1a535c4 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -6,6 +6,7 @@ authors = ["Tony Solomonik @ tony.solomonik@gmail.com"]
description = "A search engine on object storage."
repository = "https://github.com/tontinton/toshokan"
homepage = "https://github.com/tontinton/toshokan"
+readme = "./README.md"
license = "Apache-2.0 OR MIT"
[profile.release]
diff --git a/README.md b/README.md
index 980e86c..345d69a 100644
--- a/README.md
+++ b/README.md
@@ -1,3 +1,21 @@
+## Introduction
+
+`toshokan` is a search engine (think Elasticsearch, Splunk), but storing the data on object storage, most similar to Quickwit.
+
+It uses:
+* tantivy - for building and searching the inverted index data structure.
+* Apache OpenDAL - for an abstraction over object storages.
+
+I've also posted a blog post explaining the benefits and drawbacks of using an object storage for data intensive applications.
+
+## Architecture
+
+
+
+
+
+## How to use
+
```sh
toshokan create example_config.yaml
@@ -22,5 +40,8 @@ toshokan search test "tenant_id:[60 TO 65} AND severity_text:INFO" --limit 1 | j
# "timestamp": "2016-04-13T06:46:54Z"
# }
+# Merge index files for faster searching.
+toshokan merge test
+
toshokan drop test
```
diff --git a/architecture.svg b/architecture.svg
new file mode 100644
index 0000000..3eb7a73
--- /dev/null
+++ b/architecture.svg
@@ -0,0 +1 @@
+
\ No newline at end of file
+
+
+## How to use
+
```sh
toshokan create example_config.yaml
@@ -22,5 +40,8 @@ toshokan search test "tenant_id:[60 TO 65} AND severity_text:INFO" --limit 1 | j
# "timestamp": "2016-04-13T06:46:54Z"
# }
+# Merge index files for faster searching.
+toshokan merge test
+
toshokan drop test
```
diff --git a/architecture.svg b/architecture.svg
new file mode 100644
index 0000000..3eb7a73
--- /dev/null
+++ b/architecture.svg
@@ -0,0 +1 @@
+
\ No newline at end of file