I do agree with the points you brought up to the public, we should follow these steps and do it rigoursly, but that this aspect only just scratches the surface (basically the explanations are general). I would say we should do the following:

• Once we develop the SecDevOps teams, we should put them in specific groups (red, blue, green)
• The red-team is responsible for identifying the possible openings in the application (create a bounty program - you had mentioned it, but internally - and then have your blue team to see if they can circumvent the security controls the red-team put in place, then the green team should look to validate if the red or blue teams identified a real vulnerability. Have the green team look up the vulnerability to see if it exists, if not, have the green team to create tools to punish the application for one final assault or Q/A session to ensure the application does not fall to external threats or buffer overflows
• Linux - the teams should look deeper into using tools from MIT (SeLinux is one example) and determine if the tool is actually working to identify and thwart attacks (go to the "/var/log/audit/audit.log" (make sure "setenforce 1" and SeLinux is installed and working properly)
• Windows - run "get-appxprovisionedpackage -online | out-gridview -passthru | remove-appxprovisionedpackage -online" to ensure Bing applications, games and other tools are not installed, if so, remove them or utilize Uninstaller from IoBit" (group policies are good but there are HIDS like "Comodo" that are invaluable when it comes to cyber protection at the endpoint
• Look into performance and application optimization tools like CA or Extrahop to identify if the application is running at its optimal level, the tools will also help the user identify if there are holes in the development chain (utilize Agile Scrum methodologies but incorporate CMMI security procedures from Carnige Mellon, NIST and CoBIT
• Mitre At&ck is another framework we should look into when it comes to "Priviledge escalation" - https://attack.mitre.org/tactics/TA0004/, this is an area that needs attention because it delves into the methods actors use to attack a subject (again this is just one of many)

Write-Host " "

Write-Host "LSA Registry Entries"
Write-Host "--------------------"
Write-Host " "

$lsa = '{"registry":[
	{"Entry":"auditbaseobjects", "Value":"1", "Type":"Dword"},
	{"Entry":"auditbasedirectories", "Value":"1", "Type":"Dword"},
	{"Entry":"LmCompatibilityLevel", "Value":"4", "Type":"Dword"},
	{"Entry":"restrictanonymous", "Value":"1", "Type":"Dword"},
	{"Entry":"restrictanonymoussam", "Value":"1", "Type":"Dword"},
	{"Entry":"LimitBlankPasswordUse", "Value":"1", "Type":"Dword"},
    {"Entry":"SecureBoot", "Value":"1", "Type":"Dword"}
]}'

$path = "HKLM:\System\CurrentControlSet\Control\Lsa"
$regobj = ConvertFrom-Json -InputObject $lsa
$regobjects = $regobj.registry

foreach ( $i in $regobjects ) {
    $val = Get-ItemPropertyValue -Path $path -Name $i.Entry `
-ErrorAction SilentlyContinue
    if ( (Test-Path $path) -and ($i.Value -eq $val ) ) {
        Write-Host $i.Entry "registry value - ok"
    } else {
        #New-Item -path $path -Name $i.Entry -Value $i.Value -Type Dword
        $chg = Set-ItemProperty -Path $path -Name $i.Entry -Value $i.Value `
-Type $i.Type -Force
        $names = Get-ItemPropertyValue -Path $path -Name $i.Entry
        Write-Host $i.Entry "modified registry entry: " $names
    }
}