CV redaction
As hiring team, we want anonymised CVs, so we can go through a fairer and more objective screening process.
1) How can we programmatically detect things that need redacting in a CV and then how do we redact them
2) How can we do that redaction in a scalable and automated way
CV redaction
- Build vs Buy
- Word detection
- Word redaction
- Putting it all together
Build vs Buy
Word detection
- NLP (natural language processing)
- Word matching?
Word matching - regex
^[a-zA-Z0-9.!#$%&'*+/=?^_`{|}~-]+@[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?(?:\.[a-zA-Z0-9](?:[a-zA-Z0-9-]{0,61}[a-zA-Z0-9])?)*$
Word redaction
Word redaction
Tying it all together
1) How can we programmatically detect things that need redacting in a CV and then how do we redact them
2) How can we do that redaction in a scalable and automated way
✅
The infrastructure
Check it meets our restrictions on size and type
1️⃣
Scan it for viruses
2️⃣
Upload the raw file to our file store
3️⃣
Convert the file to pdf
4️⃣
Upload the pdf file to our file store
5️⃣
Redact the CV
6️⃣
Upload the redacted file to our file store
7️⃣
How (most) of the internet works
Client
💻
How (most) of the internet works
{
"email":"hewisCool@beapplied.com",
"password":"secretPassword123"
}
POST
https://app.beapplied.com/login-api
How (most) of the internet works
Server
📠
Reads request
Checks if a user with that email address
If it does checks the password is correct
If it is - ok! Return to the use
How (most) of the internet works
Client
💻
Check it meets our restrictions on size and type
1️⃣
Scan it for viruses
2️⃣
Upload the raw file to our file store
3️⃣
Convert the file to pdf
4️⃣
Upload the pdf file to our file store
5️⃣
Redact the CV
6️⃣
Upload the redacted file to our file store
7️⃣
Synchronous
Ascync
Demo