diff --git a/Cargo.lock b/Cargo.lock
index 5ea9fdc..2944fae 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -599,6 +599,16 @@ dependencies = [
  "windows-sys 0.48.0",
 ]
 
+[[package]]
+name = "storage_engine"
+version = "0.1.0"
+dependencies = [
+ "async-trait",
+ "bincode",
+ "thiserror",
+ "tokio",
+]
+
 [[package]]
 name = "strsim"
 version = "0.10.0"
diff --git a/Cargo.toml b/Cargo.toml
index 57d0219..3e6cda3 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -5,5 +5,6 @@ members = [
     "proto",
     "server",
     "client",
-    "parser"
+    "parser",
+    "storage_engine"
 ]
diff --git a/storage_engine/Cargo.toml b/storage_engine/Cargo.toml
new file mode 100644
index 0000000..6658764
--- /dev/null
+++ b/storage_engine/Cargo.toml
@@ -0,0 +1,10 @@
+[package]
+name = "storage_engine"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+bincode = "2.0.0-rc.3"
+tokio = { version = "1.34.0", features = ["full"] }
+async-trait = "0.1.74"
+thiserror = "1.0.50"
diff --git a/storage_engine/src/lib.rs b/storage_engine/src/lib.rs
new file mode 100644
index 0000000..0e2eb79
--- /dev/null
+++ b/storage_engine/src/lib.rs
@@ -0,0 +1,2 @@
+
+pub mod storage_engine;
diff --git a/storage_engine/src/main.rs b/storage_engine/src/main.rs
new file mode 100644
index 0000000..fffde93
--- /dev/null
+++ b/storage_engine/src/main.rs
@@ -0,0 +1,54 @@
+use tokio::sync::{Mutex, RwLock};
+use tokio::fs::{File, OpenOptions};
+use tokio::io::{BufReader, BufWriter, AsyncReadExt, AsyncWriteExt, AsyncSeekExt, SeekFrom};
+use tokio::fs;
+
+mod storage_engine;
+
+use crate::storage_engine::*;
+
+#[tokio::main]
+async fn main() -> Result<(), std::io::Error> {
+    println!("STOOOOOOOOOOOORAAAAAAAAAAAGE");
+
+    let blob_name = "blob10.minisql";
+
+    // WARNING: Number of columns is 5?????
+
+    let mut store = Store::new(blob_name, 5).await.map_err(|e| e.to_io_or_panic())?;
+
+    // let store_bytes = store.get_all_bytes().await.map_err(|e| e.to_io_or_panic())?;
+    // println!("{:?}", store_bytes);
+
+    let mut buff: Vec<u8> = vec![0;1];
+    let x = store.file.read_exact(&mut buff[..]).await?;
+    println!("{:?}", buff);
+
+
+    // let entry0: Entry<u32> = Entry::new_deleted(vec![1, 2, 3, 4, 5]);
+    // let entry1: Entry<u32> = Entry::new_deleted(vec![200,200,5,6,7]);
+    // let cursor0 = store.append_entry(&entry0).await.map_err(|e| e.to_io_or_panic())?;
+    // // println!("cursor0 = {}", cursor0);
+
+    // let cursor1 = store.append_entry(&entry1).await.map_err(|e| e.to_io_or_panic())?;
+    // println!("cursor0 = {}, cursor1 = {}", cursor0, cursor1);
+
+    // let mut store = Store::connect(blob_name).await.map_err(|e| e.to_io_or_panic())?;
+    // println!("{:?}", store);
+
+    // let x = store.entry_at::<u32>(16).await.map_err(|e| e.to_io_or_panic())?;
+    // println!("{:?}", x);
+
+    // let store_bytes = store.get_all_bytes().await.map_err(|e| e.to_io_or_panic())?;
+    // println!("{:?}", store_bytes);
+
+    // let mut store = ColumnStore::connect("blob08.minisql").await.map_err(|e| e.to_io_or_panic())?;
+    // let entry2: StoreEntry<u32> = StoreEntry::new_deleted(vec![3, 2, 1]);
+    // let cursor2 = store.append_entry(&entry2).await.map_err(|e| e.to_io_or_panic())?;
+    // println!("cursor2 = {}", cursor2);
+    // println!("{:?}", store);
+
+
+    println!("DONE");
+    Ok(())
+}
diff --git a/storage_engine/src/storage_engine.rs b/storage_engine/src/storage_engine.rs
new file mode 100644
index 0000000..802444e
--- /dev/null
+++ b/storage_engine/src/storage_engine.rs
@@ -0,0 +1,436 @@
+use tokio::io::{BufReader, BufWriter, AsyncReadExt, AsyncWriteExt, AsyncSeekExt, SeekFrom};
+use tokio::sync::{Mutex, RwLock};
+use tokio::fs::{File, OpenOptions};
+use tokio::fs;
+
+use bincode;
+use bincode::de::Decoder;
+use bincode::enc::write::Writer;
+use bincode::enc::Encoder;
+use bincode::{BorrowDecode, Decode, Encode};
+use bincode::config::{BigEndian, Configuration, Fixint};
+
+use std::mem::size_of;
+
+
+// =============Byte encoding/decoding============ 
+const BIN_CONFIG: Configuration<BigEndian, Fixint> = bincode::config::standard().with_big_endian().with_fixed_int_encoding();
+
+fn encode<T: Encode>(t: &T) -> Result<Vec<u8>, bincode::error::EncodeError> {
+    bincode::encode_to_vec(t, BIN_CONFIG)
+}
+
+fn decode<T: Decode>(bytes: &[u8]) -> Result<(T, usize), bincode::error::DecodeError> {
+    bincode::decode_from_slice(bytes, BIN_CONFIG)
+}
+
+fn encode_vector<T: Encode>(ts: &[T]) -> Result<Vec<u8>, bincode::error::EncodeError> {
+    let size: usize = ts.len();
+    let mut result = encode(&size)?;
+    for t in ts {
+        result.append(&mut encode(&t)?);
+    }
+    Ok(result)
+}
+
+fn decode_vector<T: Decode>(bytes: &[u8]) -> Result<Vec<T>, bincode::error::DecodeError> {
+    let mut offset = size_of::<usize>();
+    let result_len: usize = decode(&bytes[..offset])?.0;
+    let mut result: Vec<T> = Vec::with_capacity(result_len);
+    for _ in 0..result_len {
+        let (x, bytes_consumed) = decode::<T>(&bytes[offset..])?;
+        offset += bytes_consumed;
+        result.push(x);
+    }
+    Ok(result)
+}
+
+// We don't care about encoding the length here (since it will be used for a row with known column
+// size)
+fn encode_sequence<T: Encode>(ts: &[T]) -> Result<Vec<u8>, bincode::error::EncodeError> {
+    let mut result = vec![];
+    for t in ts {
+        result.append(&mut encode(&t)?);
+
+    }
+    Ok(result)
+}
+
+fn decode_sequence<T: Decode>(len: usize, bytes: &[u8]) -> Result<Vec<T>, bincode::error::DecodeError> {
+    let mut result: Vec<T> = Vec::with_capacity(len);
+    let mut offset = 0;
+    for _ in 0..len {
+        let (x, bytes_consumed) = decode::<T>(&bytes[offset..])?;
+        offset += bytes_consumed;
+        result.push(x);
+    }
+    Ok(result)
+}
+
+
+fn example_encoding_decoding() {
+    let xs: Vec<u32> = vec![123, 250, 256, 123, 123, 123];
+    let xs: Vec<u32> = vec![];
+    let xs: Vec<u32> = vec![123];
+    let xs: Vec<u32> = vec![123, 250];
+
+    let xs: Vec<String> = vec!["foo".to_string(), "bar".to_string()];
+
+
+    println!("original {:?}", xs);
+    let exs = encode_vector(&xs[..]).unwrap();
+    println!("encoded {:?}", exs);
+
+    // WARNING: Don't forget to specify the type here
+    // let dxs = decode_vector::<u32>(&exs[..]).unwrap();
+    let dxs = decode_vector::<String>(&exs[..]).unwrap();
+    println!("decoded {:?}", dxs);
+}
+
+// ============Column Store===============
+
+// ColumnStore
+#[derive(Debug)]
+pub struct Store {
+    column_file_name: String,
+    // TODO: This should be private
+    pub file: File,
+    header: StoreHeader
+    // meta
+    //   location of rows file
+    //   locations of index files
+    //     
+    // rows file
+    // list
+}
+
+// These type aliases are here because they make writing decoders easier.
+type NumberOfColumns = usize;
+type DeletedCount = usize;
+#[derive(Debug)]
+pub struct StoreHeader {
+    number_of_columns: NumberOfColumns,
+    deleted_count: DeletedCount,
+}
+
+#[derive(Debug)]
+pub struct Entry<T> {
+    is_deleted: bool,
+    // file_position: FilePosition,
+    data: Vec<T>,
+}
+
+#[derive(Debug)]
+pub enum Error {
+    DecodeError(DecodeErrorKind, bincode::error::DecodeError),
+    EncodeError(bincode::error::EncodeError),
+    IoError(std::io::Error),
+    InvalidStoreHeader,
+}
+
+#[derive(Debug)]
+pub enum DecodeErrorKind {
+    StoreHeaderNumberOfColumns,
+    StoreHeaderDeletedCount,
+    EntryData,
+    EntryIsDeleted,
+    EntryDataSize
+}
+
+// ===Errors===
+impl Error {
+    pub fn to_io_or_panic(self) -> std::io::Error {
+        use Error::*;
+        match self {
+            IoError(err) => err,
+            err => {
+                println!("{:?}", err);
+                panic!();
+            }
+        }
+    }
+}
+
+impl From<bincode::error::EncodeError> for Error {
+    fn from(err: bincode::error::EncodeError) -> Self {
+        Self::EncodeError(err)
+    }
+}
+
+impl From<std::io::Error> for Error {
+    fn from(err: std::io::Error) -> Self {
+        Self::IoError(err)
+    }
+}
+
+
+// ====Entry====
+impl <T>Entry<T> {
+    pub fn new(data: Vec<T>) -> Self {
+        Self { is_deleted: false, data }
+    }
+
+    pub fn new_deleted(data: Vec<T>) -> Self {
+        Self { is_deleted: true, data }
+    }
+
+    // FORMAT: [HEADER, ..sequence of data]
+    //     HEADER: [Boolean (one byte), number of bytes in the data (not including the boolean)]
+    fn encode(self: &Entry<T>) -> Result<Vec<u8>, Error> 
+        where T: Encode
+    {
+        let mut result: Vec<u8> = encode(&self.is_deleted)?; // bool 1 byte
+        let mut encoded_data = encode_sequence(&self.data[..])?;
+        let encoded_data_len = encoded_data.len();
+        result.append(&mut encode(&encoded_data_len)?); // usize 8 bytes
+        println!("enc data len == {}", encoded_data_len);
+        println!("encoded_data == {:?} ", encoded_data);
+        result.append(&mut encoded_data); // data variable size
+        Ok(result)
+    }
+
+    // in bytes
+    pub fn header_size() -> usize {
+        size_of::<bool>() + size_of::<usize>()
+    }
+
+    // TODO: Maybe introduce an EntryHeader as a separate type?
+    pub fn decode_header(header_bytes: Vec<u8>) -> Result<(bool, usize), Error> {
+        let (is_deleted, offset) =
+            decode::<bool>(&header_bytes)
+                .map_err(|e| Error::DecodeError(DecodeErrorKind::EntryIsDeleted, e))?;
+
+        let (data_size, _) = 
+            decode::<usize>(&header_bytes[offset..])
+                .map_err(|e| Error::DecodeError(DecodeErrorKind::EntryDataSize, e))?;
+        Ok((is_deleted, data_size))
+    }
+}
+
+pub type Column = u64;
+pub type Cursor = u64;
+
+impl StoreHeader {
+    fn encode(&self) -> Result<Vec<u8>, Error> {
+        // FORMAT: First Number of Columns, Then Deleted Count.
+        let mut result = encode(&self.number_of_columns)?;
+        result.append(&mut encode(&self.deleted_count)?);
+        Ok(result)
+    }
+
+    async fn decode(file: &mut File) -> Result<StoreHeader, Error> {
+        let number_of_columns_size = size_of::<NumberOfColumns>();
+        let deleted_count_size = size_of::<DeletedCount>();
+        let header_size: usize = number_of_columns_size + deleted_count_size;
+        let mut header_bytes: Vec<u8> = vec![0; header_size];
+        // TODO: Why do we need to have mutable reference for the file when we are reading it?
+        match file.read_exact(&mut header_bytes).await {
+            Ok(_) => {
+                let offset = 0;
+                let (number_of_columns, offset) =
+                    decode::<NumberOfColumns>(&header_bytes[offset..offset + number_of_columns_size])
+                        .map_err(|e| Error::DecodeError(DecodeErrorKind::StoreHeaderNumberOfColumns, e))?;
+                let (deleted_count, _) =
+                    decode::<DeletedCount>(&header_bytes[offset..offset + deleted_count_size])
+                        .map_err(|e| Error::DecodeError(DecodeErrorKind::StoreHeaderDeletedCount, e))?;
+                let header = StoreHeader {
+                    number_of_columns,
+                    deleted_count,
+                };
+                Ok(header)
+            },
+            Err(err) => {
+                // TODO: When err is of the kind UnexpectedEof, return InvalidStoreHeader
+                println!("ARE WE HERE?");
+                Err(Error::from(err))
+            }
+        }
+    }
+}
+
+
+impl Store {
+    // For debugging
+    pub async fn get_all_bytes(mut self) -> Result<Vec<u8>, Error>{
+        let mut bytes: Vec<u8> = vec![];
+        use std::io::Read;
+        // for byte in self.file.bytes() {
+        // }
+        self.file.read_exact(&mut bytes[..]).await?;
+        Ok(bytes)
+    }
+
+    pub async fn new(column_file_name: &str, number_of_columns: usize) -> Result<Self, Error> {
+        let mut file =
+            OpenOptions::new()
+                .write(true)
+                .read(true)
+                .create_new(true)
+                .open(column_file_name)
+                .await?;
+
+        let header = StoreHeader {
+            number_of_columns,
+            deleted_count: 0,
+        } ;
+        let encoded_header = header.encode()?;
+        file.write(&encoded_header).await?;
+        println!("is something being encoded? {:?}", encoded_header);
+        let store = Self {
+            column_file_name: column_file_name.to_string(),
+            file,
+            header,
+        };
+        Ok(store)
+    }
+
+    pub async fn connect(column_file_name: &str) -> Result<Self, Error> {
+        let mut file = OpenOptions::new().read(true).write(true).open(column_file_name).await?;
+
+        let header = StoreHeader::decode(&mut file).await?;
+        Ok(Self {
+            column_file_name: column_file_name.to_string(),
+            file,
+            header
+        })
+    }
+
+    pub async fn entry_at<T: Decode>(&mut self, cursor: Cursor) -> Result<Entry<T>, Error> {
+        self.file.seek(SeekFrom::Start(cursor)).await?;
+
+        // 1. read header bytes (fixed number of bytes).
+        // 2. decode header
+        // 3. read entry data bytes.
+        // 4. decode data
+        // That will tell us how much data there is.
+        let entry_header_size = Entry::<T>::header_size();
+        let mut header_bytes: Vec<u8> = vec![0; entry_header_size];
+        self.file.read_exact(&mut header_bytes).await?;
+
+        println!("cursor == {}", cursor);
+        println!("header_bytes == {:?}", header_bytes);
+
+        let (is_deleted, data_size) = Entry::<T>::decode_header(header_bytes)?;
+
+        self.file.seek(SeekFrom::Current(entry_header_size as i64)).await?;
+        let mut data_bytes: Vec<u8> = vec![0; data_size];
+
+        println!("(is_delted, data_size) = ({}, {})", is_deleted, data_size);
+        let data =
+            decode_sequence::<T>(self.header.number_of_columns, &mut data_bytes)
+                .map_err(|e| Error::DecodeError(DecodeErrorKind::EntryData, e))?;
+        Ok(Entry {
+            is_deleted,
+            data
+        })
+    }
+
+    pub async fn append_entry<T: Encode>(&mut self, entry: &Entry<T>) -> Result<Cursor, Error> {
+        // On linux when opening a file in append mode, the seek is set to 0
+        // and only updated after a write. That's why we do the cursor gymnastic at the end.
+        let encoded_entry: Vec<u8> = entry.encode()?;
+        println!("encoded_entry == {:?}", encoded_entry);
+        println!("bool size == {}", size_of::<bool>());
+        println!("usize size == {}", size_of::<usize>());
+        self.file.write(&encoded_entry).await?;
+        let next_cursor: Cursor = self.file.stream_position().await?;
+        let cursor: Cursor = next_cursor - encoded_entry.len() as u64;
+        Ok(cursor)
+    }
+
+    pub async fn iterate_all<T>(&mut self) -> Result<Cursor, Error> {
+        // Loop through the rows and print them out
+        todo!()
+    }
+}
+
+// impl StorageEngine for ColumnStore {
+//     async fn append(&mut self, id: Index, entry: Row<T>) -> Result<???, Error>
+
+//     async fn get_all(&self) -> ???
+//     async fn get_eq(&self, column: Column, value: T) -> ???
+
+//     async fn delete_all(&mut self)
+//     async fn delete_eq(&mut self, column: Column, value: T) -> ???
+// }
+
+// struct Error {
+// }
+
+
+// Selected(
+//     &'a TableSchema,
+//     ColumnSelection,
+//     TODO: Don't do the Box(dyn Iterator<...>)
+//       you'll have a concrete implementation of Iterator, and that's what you'll use
+//     Box<dyn Iterator<Item = RestrictedRow> + 'a + Send>,
+// ),
+
+
+
+// #[async_trait]
+// trait StorageEngine<T>
+// where T: Encode + Decode
+// {
+//     async fn append(&mut self, id: Index, entry: Row<T>) -> Result<???, Error>
+
+//     async fn get_all(&self) -> ???
+//     async fn get_eq(&self, column: Column, value: T) -> ???
+
+//     async fn delete_all(&mut self)
+//     async fn delete_eq(&mut self, column: Column, value: T) -> ???
+// }
+
+// #[cfg(test)]
+// mod tests {
+//     #[test]
+//     fn hello_test() {
+//         assert!(true);
+//     }
+// }
+
+// let sroage_engine = STorageEngine::new("users")
+// let mut next_position = 0
+
+
+// type FilePosition = usize;
+
+
+// type StoreFile = Vec<Entry>;
+// type IndexFile = ???
+
+// struct IndexEntry {
+
+// }
+
+
+// #00000 [false, u26, "Arnold", "schwarzenegger", "gettothechoppa@yahoo.com"]  #5120000 [true, u27, "Arnold", "Vosloo", "avosloo@aol.com"]
+// #00000 [true, u27, "Arnold", "Vosloo", "avosloo@aol.com"]
+
+
+// at #00000  512 kb deleted, 
+// ...
+
+
+
+
+// [(u26, [#00000]), (u27, [#5120000])]
+// [("Arnold", [#000000, #5120000]), ("Arnfsdaf", []), ("Adasdsd", []), ("Bdsad", [])]
+// // basically always keep indexes in memory and on write always sync on disk
+
+
+
+
+// CREATE INDEX usersname ON "users" (name);
+
+// INSERT INTO users (id, name, surname, email) VALUES (u26, "Arnold", "schwarzenegger", "gettothechoppa@yahoo.com");
+// INSERT INTO users (id, name, surname, email) VALUES (u27, "Arnold", "Vosloo", "avosloo@aol.com");
+
+
+// SELECT * FROM users WHERE id=u26;
+
+// SELECT * FROM users WHERE name="Arnold";
+
+
+// SELECT * FROM cars;
+// DELETE FROM users WHERE name="Arnold";