System Admin required for Ubuntu / AWS CLI data offloading (backup) solution

I need a GUN Ubuntu System Admin with AWESOME AWS expertise. You are probably going to be a Centos freak as well - and I'll appreciate that too (for potential ongoing work). If you don't know how to write insane scripts and you are not all over the AWS CLI then DONT. BOTHER. PITCHING. If you don't know what IAM Users are and how to craft clean policies, then log off freelancer now and consider your life choices. If you can collaborate with me during 11am to 6pm ADST/AEST, then please make that clear in your pitch. (I am based in Melbourne and prefer Slack to communicate and collaborate in real-time - does anyone email anymore?). I run a managed services agency, and I want to update the way we perform disaster recovery offsite data storage. This will be developed on one production server, and we will roll it out to the remaining servers. I need a script, or set of scripts that will iterate through a configured set of directories and do the following: target_directory=/home/user1/:/home/user2 backup_staging=/backups s3_bucket=s3://bucket server_ref=prod01 notification_email=myticketingsystem@mydomain. [login to view URL] Iterarate through all subdirectories in target_directory and create a corresponding [login to view URL] archive including date/time in the filename in backup_staging. For example, /users/fred/www.fred. com would be archived to /backups/fred/www.fred. com/www.fred. [login to view URL] During the archiving process, the .git directory in the subdirectory root MUST NOT be included. Once the archive is created, the script should use AWS CLI to copy the archive to the bucket, creating a directory structure if required to mirror inside the backups directory but using the server_ref first . i.e. s3://bucket/server_ref/[login to view URL][archive] Once the archive is sent to S3, any archives older than 45 days should be removed from S3. (this can be done via policy if possible/preferred). Once the archive is CONFIRMED to have been copied to S3, the local archive should be removed - ie /backups/fred/www.fred. com/www.fred. [login to view URL] The script should then process the NEXT subdirectory in the target directory. This means one archive is created, offloaded and removed from the staging area before the next one is processed. Once the first target_dir is processed, the next one should be processed. If an error occurs at any time, an email should be sent. Once the offloading is complete for all sites, an email should be sent. The script should report on how long it took to complete when reporting completion email. The script should be set to run at 2am each morning. You should tell me what is good/bad/indifferent about my approach (bearing in mind this will be running on 16 servers) and help me craft something even better (if that's at all possible). Stage 2 of this will be to configure a VM (somewhere) with x TB storage to use AWS CLI Sync to copy down ALL server backups from the S3 storage to a local storage device. Tell me in your pitch if you know whether "aws sync" deletes files in the destination location or reflects changes in the destination back on the source. You will need to sign an NDA before starting on this. Also, let me know your favourite brand of coffee and what you thought of The Expanse.