diff --git a/data_management/README.md b/data_management/README.md
index 2e0c258..8ce1a99 100644
--- a/data_management/README.md
+++ b/data_management/README.md
@@ -114,5 +114,29 @@ This endpoint retrieves the packages that are ready to be moved from Globus. The
 ### rebuild library modules on change
 `$ python3 setup.py install --user`
 
+## DLU Watcher
+
+The dlu_watcher is controlled via the DluWatcher docker file and runs watch_files.py. It uses many of the same services as are used for the data-manager-service. The dlu watcher does a lot of background processing of data that is often triggered by the data managers updating records in the DMD tables.
+
+### Slide Renaming
+One major (new) functionality of the dlu-watcher is the slide renaming process. We periodically check the slide_manifest_import table for new records that we have not put into the slide_scan_curation table (this is a pretty brain-dead check of equal number of rows). If there are new rows in the slide_manifest_import table, we continue processing those rows.
+
+The majority of the functionality for processing slide renaming lives in slide_management.py in services.
+
+The first step is to process the new rows in slide_manifest_import and verify that we have all of the neccesary information in order to do an import. If the process is not picking up new records in the slide_manifest_import, check the log files in dlu_watcher for potential errors as we have not created a record in the slide_scan_curation table at this point, and do not want to since there are issues.
+
+If records meet the preliminary checks, then they get added to the slide_scan_curation table. From here on, if there is an error, we will add that error to the appropriate records inside of slide_scan_curation.
+
+### Moving files to DataLake
+The other function of the dlu-watcher is to move files from Globus into the Data Lake. There are 2 main paths here.
+
+#### Whole Slide Images 
+
+For Whole slide images that are not a recalled package, we need to go down a different path to move the files into the data lake. We need to rename the files and then move the files in place and update the databases. This is a parallel path to non-WSI files, and calls many of the same functions, but ended up needing to be a partial duplicate of the process for non-wsi, because we were unable to rename the files on copy.
+
+#### Non-WSI 
+
+This will do a number of checks to make sure everything is copasetic and then call out to the appropriate locations to create new records in the databases, create directories in the data lake, and copy the data into the data lake. This also includes generating the md5checksums to store in the databases and can sometimes cause issues when we have extremely large files.
+
 ## Known Bug
 There is a known [bug with docker on MacOS](https://github.com/docker/for-mac/issues/2670) in which the container is unable to talk to the host network. This problem may occur when attempting to connect to a tunnel created on the host machine. To work around this issue, you can either run this on a linux machine/windows machine, or bypass docker completely and run the script directly on your local machine.