diff --git a/README.md b/README.md
index c6d94f4..dce4e3d 100644
--- a/README.md
+++ b/README.md
@@ -174,10 +174,60 @@ The BinaryTable uses a custom binary format optimized for random access:
 - **Concurrent Access**: Single-threaded access only
 - **Type System**: Limited compared to JSON (no nested objects, mixed arrays)
 
-## Future Roadmap
+# SweepStore Roadmap
 
-- Full JSON parity with nested objects and mixed-type arrays
-- Compression support for large data blocks
-- Checksums and data integrity validation
-- Multi-threaded access with file locking
-- Query system for complex data retrieval
\ No newline at end of file
+## v1.0.0 (Current)
+- ✅ Key-value storage with hash-based indexing
+- ✅ Uniform arrays with random access
+- ✅ Automatic memory management via free list
+- ✅ Single file format (.sws)
+
+## v2.0.0 (Planned)
+
+### Core Changes
+- **Nested objects** - Store objects within objects for hierarchical data
+- **Key enumeration** - Query and list all keys without knowing them upfront
+- **Object-based architecture** - Each object has its own address table and key list
+
+### File Format Changes
+```
+v1: Global address table → All keys at root level
+v2: Root object → Each object manages its own keys
+```
+
+### New API
+```dart
+// Nested objects
+table["player"] = {};
+table["player"]["name"] = "Alice";
+table["player"]["inventory"] = {};
+
+// Key discovery
+List<String> keys = table.keys();
+bool exists = table.containsKey("player");
+
+// Object operations
+BT_Object player = table["player"];
+for (String key in player.keys()) {
+  print("$key: ${player[key]}");
+}
+```
+
+### Implementation Strategy
+- Introduce `BT_Object` type with embedded address table + key list
+- `BinaryTable` becomes wrapper around root object
+- Maintain global free list (not per-object)
+- Address table entries remain absolute pointers
+- Key list stored alongside each object's address table
+
+### Breaking Changes
+- File format incompatible with v1.0.0
+- API remains largely the same (only additions)
+- Migration tool needed for v1 → v2 files
+
+## Future Considerations
+- Hash collision resolution
+- Checksums for data integrity
+- Optimized allocation strategies
+- Incremental array resizing
+- Compression options
\ No newline at end of file